Neuroeconomics How Neuroscience Can Inform Economics

Journal of Economic LiteratureVol. XLIII (March 2005), pp. 9–64

Neuroeconomics: How NeuroscienceCan Inform Economics

COLIN CAMERER, GEORGE LOEWENSTEIN, and DRAZEN PRELEC∗

Who knows what I want to do? Who knows what anyone wants to do? How can yoube sure about something like that? Isn’t it all a question of brain chemistry, signalsgoing back and forth, electrical energy in the cortex? How do you know whethersomething is really what you want to do or just some kind of nerve impulse in thebrain. Some minor little activity takes place somewhere in this unimportant place inone of the brain hemispheres and suddenly I want to go to Montana or I don’t wantto go to Montana. (White Noise, Don DeLillo)

9

∗ Camerer: California Institute of Technology.Loewenstein: Carnegie Mellon University. Prelec:Massachusetts Institute of Technology. We thank partici-pants at the Russell Sage Foundation-sponsored confer-ence on Neurobehavioral Economics (May 1997) atCarnegie Mellon, the Princeton workshop on NeuralEconomics December 8–9, 2000, and the Arizona confer-ence in March 2001. This research was supported by NSFgrant SBR-9601236 and by the Center for Advanced Studyin Behavioral Sciences, where the authors visited during1997–98. David Laibson’s presentations have been partic-ularly helpful, as were comments and suggestions from theeditor and referees, and conversations and comments fromRalph Adolphs, John Allman, Warren Bickel, Greg Berns,Meghana Bhatt, Jonathan Cohen, Angus Deaton, JohnDickhaut, Paul Glimcher, Dave Grether, Ming Hsu, DavidLaibson, Danica Mijovic-Prelec, Read Montague, CharliePlott, Matthew Rabin, Antonio Rangel, Peter Shizgal,Steve Quartz, and Paul Zak. Albert Bollard, EstherHwang, and Karen Kerbs provided editorial assistance.

1. Introduction

In the last two decades, following almost acentury of separation, economics has begunto import insights from psychology.“Behavioral economics” is now a prominentfixture on the intellectual landscape and hasspawned applications to topics in economics,

such as finance, game theory, labor econom-ics, public finance, law, and macroeconomics(see Colin Camerer and George Loewenstein2004). Behavioral economics has mostly beeninformed by a branch of psychology called“behavioral decision research,” but othercognitive sciences are ripe for harvest. Someimportant insights will surely come from neu-roscience, either directly or because neuro-science will reshape what is believed aboutpsychology which in turn informs economics.

Neuroscience uses imaging of brain activityand other techniques to infer details abouthow the brain works. The brain is the ultimate“black box.” The foundations of economictheory were constructed assuming that detailsabout the functioning of the brain’s black boxwould not be known. This pessimism wasexpressed by William Jevons in 1871:

I hesitate to say that men will ever have the meansof measuring directly the feelings of the humanheart. It is from the quantitative effects of thefeelings that we must estimate their comparativeamounts.

mr05_Article 1 3/28/05 3:25 PM Page 9

10 Journal of Economic Literature, Vol. XLIII (March 2005)

Since feelings were meant to predict behav-ior but could only be assessed from behavior,economists realized that, without directmeasurement, feelings were useless inter-vening constructs. In the 1940s, the con-cepts of ordinal utility and revealedpreference eliminated the superfluous inter-mediate step of positing immeasurable feel-ings. Revealed preference theory simplyequates unobserved preferences withobserved choices. Circularity is avoided byassuming that people behave consistently,which makes the theory falsifiable; oncethey have revealed that they prefer A to B,people should not subsequently choose Bover A. Later extensions—discounted,expected, and subjective expected utility,and Bayesian updating—provided similar“as if” tools which sidestepped psychologicaldetail. The “as if” approach made goodsense as long as the brain remained substan-tially a black box. The development of eco-nomics could not be held hostage toprogress in other human sciences.

But now neuroscience has provedJevons’s pessimistic prediction wrong; thestudy of the brain and nervous system isbeginning to allow direct measurement ofthoughts and feelings. These measurementsare, in turn, challenging our understandingof the relation between mind and action,leading to new theoretical constructs andcalling old ones into question. How can thenew findings of neuroscience, and the theo-ries they have spawned, inform an econom-ic theory that developed so impressively intheir absence?

In thinking about the ways that neuro-science can inform economics, it is useful todistinguish two types of contributions, whichwe term incremental and radical approach-es. In the incremental approach, neuro-science adds variables to conventionalaccounts of decision making or suggests spe-cific functional forms to replace “as if ”assumptions that have never been well sup-ported empirically. For example, research onthe neurobiology of addiction suggests how

drug consumption limits pleasure fromfuture consumption of other goods (dynam-ic cross-partial effects in utility for commod-ity bundles) and how environmental cuestrigger unpleasant craving and increasedemand. These effects can be approximatedby extending standard theory and thenapplying conventional tools (see DouglasBernheim and Antonio Rangel 2004; DavidLaibson 2001; Ted O’Donoghue andMatthew Rabin 1997).

The radical approach involves turningback the hands of time and asking how eco-nomics might have evolved differently if ithad been informed from the start by insightsand findings now available from neuro-science. Neuroscience, we will argue, pointsto an entirely new set of constructs to under-lie economic decision making. The standardeconomic theory of constrained utility maxi-mization is most naturally interpreted eitheras the result of learning based on consump-tion experiences (which is of little help whenprices, income, and opportunity setschange), or careful deliberation—a balanc-ing of the costs and benefits of differentoptions—as might characterize complexdecisions like planning for retirement, buy-ing a house, or hammering out a contract.Although economists may privately acknowl-edge that actual flesh-and-blood humanbeings often choose without much delibera-tion, the economic models as written invari-ably represent decisions in a “deliberativeequilibrium,” i.e., that are at a stage wherefurther deliberation, computation, reflec-tion, etc. would not by itself alter the agent’schoice. The variables that enter into the for-mulation of the decision problem—the pref-erences, information, and constraints—areprecisely the variables that should affect thedecision, if the person had unlimited timeand computing ability.

While not denying that deliberation is partof human decision making, neurosciencepoints out two generic inadequacies of thisapproach—its inability to handle the crucialroles of automatic and emotional processing.


Camerer, Loewenstein, and Prelec: Neuroeconomics 11

First, much of the brain implements“automatic” processes, which are faster thanconscious deliberations and which occurwith little or no awareness or feeling of effort(John Bargh et al. 1996; Bargh and TanyaChartrand 1999; Walter Schneider andRichard Shiffrin 1977; Shiffrin andSchneider 1977). Because people have littleor no introspective access to these processes,or volitional control over them, and theseprocesses were evolved to solve problems ofevolutionary importance rather than respectlogical dicta, the behavior these processesgenerate need not follow normative axiomsof inference and choice.

Second, our behavior is strongly influ-enced by finely tuned affective (emotion)systems whose basic design is common tohumans and many animals (Joseph LeDoux1996; Jaak Panksepp 1998; Edmund Rolls1999). These systems are essential for dailyfunctioning, and when they are damaged orperturbed, by brain injury, stress, imbal-ances in neurotransmitters, or the “heat ofthe moment,” the logical-deliberative sys-tem—even if completely intact—cannotregulate behavior appropriately.

Human behavior thus requires a fluidinteraction between controlled and automat-ic processes, and between cognitive andaffective systems. However, many behaviorsthat emerge from this interplay are routine-ly and falsely interpreted as being the prod-uct of cognitive deliberation alone (GeorgeWolford, Michael Miller, and MichaelGazzaniga 2000). These results (some ofwhich are described below) suggest thatintrospective accounts of the basis for choiceshould be taken with a grain of salt. Becauseautomatic processes are designed to keepbehavior “off-line” and below consciousness,we have far more introspective access tocontrolled than to automatic processes.Since we see only the top of the automaticiceberg, we naturally tend to exaggerate theimportance of control.

Neuroscience findings and methods willundoubtedly play an increasingly prominent

1 The first meeting was held at Carnegie Mellon in1997. Later meetings were held in Arizona andPrinceton, in 2001, in Minnesota in 2002, and onMartha’s Vineyard in 2003 and Kiawah Island in 2004.Occasional sessions devoted to this rapidly growing topicare now common at large annual meetings in both economics and neuroscience.

role in economics and other social sciences(e.g., law, Terrence Chorvat, Kevin McCabe,and Vernon Smith 2004). Indeed, a new areaof economics that has been branded “neu-roeconomics” has already formed the basisof numerous academic gatherings that havebrought neuroscientists and economiststogether (see also Paul Zak, in press).1

Participating in the development of a sharedintellectual enterprise will help us ensurethat the neuroscience informs economicquestions we care about. Stimulated by theauthors’ own participation in a number ofsuch meetings, our goal in this paper is todescribe what neuroscientists do and howtheir discoveries and views of human behav-ior might inform economic analysis. In thenext section (2), we describe the diversity oftools that neuroscientists use. Section 3introduces a simplified account of how cog-nition and affect on the one hand, and auto-matic and controlled processes on the other,work separately, and interact. Section 4 dis-cusses general implications for economics.Section 5 goes into greater detail aboutimplications of neuroeconomics for four top-ics in economics: intertemporal choice, deci-sion making under risk, game theory, andlabor-market discrimination. Through mostof the paper, our focus is largely on how neu-roscience can inform models of microfoun-dations of individual decision making.Section 6 discusses some broader macroimplications and concludes.

2. Neuroscience Methods

Scientific technologies are not just toolsscientists use to explore areas of interest.New tools also define new scientific fieldsand erase old boundaries. The telescope cre-ated astronomy by elevating the science from



pure cosmological speculation. The micro-scope made possible similar advances in biol-ogy. The same is true of economics. Itsboundaries have been constantly reshaped bytools such as mathematical, econometric, andsimulation methods. Likewise, the currentsurge of interest in neuroscience by psychol-ogists emerged largely from new methods,and the methods may productively blur theboundaries of economics and psychology.This section reviews some of these methods.

2.1 Brain Imaging

Brain imaging is currently the most popu-lar neuroscientific tool. Most brain imaginginvolves a comparison of people performingdifferent tasks—an “experimental” task anda “control” task. The difference betweenimages taken while subject is performing thetwo tasks provides a picture of regions of thebrain that are differentially activated by theexperimental task.

There are three basic imaging methods.The oldest, electro-encephalogram (or EEG)uses electrodes attached to the scalp to meas-ure electrical activity synchronized to stimu-lus events or behavioral responses (known asEvent Related Potentials or ERPs). LikeEEG, positron emission topography (PET)scanning is an old technique in the rapidlychanging time-frame of neuroscience, but isstill a useful technique. PET measures bloodflow in the brain, which is a reasonable proxyfor neural activity, since neural activity in aregion leads to increased blood flow to thatregion. The newest, and currently most pop-ular, imaging method is functional magneticresonance imaging (fMRI), which tracksblood flow in the brain using changes in mag-netic properties due to blood oxygenation(the “BOLD signal”). Simultaneous directrecording of neural processing and fMRIresponses confirms that the BOLD signalreflects input to neurons and their processing(Nikos Logothetis et al. 2001).

Although fMRI is increasingly becomingthe method of choice, each method has pros and cons. EEG has excellent temporal

resolution (on the order of one millisecond)and is the only method used with humansthat directly monitors neural activity, asopposed to, e.g., blood flow. But spatial reso-lution is poor and it can only measure activi-ty in the outer part of the brain. EEGresolution has, however, been improvingthrough the use of ever-increasing numbersof electrodes. Interpolation methods, and thecombined use of EEG and fMRI for measur-ing outer-brain signals and inner-brain sig-nals at the same time, promise to createstatistical methods which make reasonableinferences about activity throughout thebrain from EEG signals. For economics, amajor advantage of EEG is its relative unob-trusiveness and portability. Portability willeventually reach the point where it will bepossible to take unobtrusive measurementsfrom people as they go about their dailyaffairs. PET and fMRI provide better spatialresolution than EEG, but poorer temporalresolution because blood-flow to neurallyactive areas occurs with a stochastic lag froma few seconds (fMRI) to a minute (PET).

Brain imaging still provides only a crudesnapshot of brain activity. Neural processesare thought to occur on a 0.1 millimeterscale in 100 milliseconds (msec), but thespatial and temporal resolution of a typicalscanner is only 3 millimeters and several sec-onds. Multiple trials per subject can be aver-aged to form composite images, but doing soconstrains experimental design. However,the technology has improved rapidly and willcontinue to improve. Hybrid techniques thatcombine the strengths of different methodsare particularly promising. Techniques forsimultaneously scanning multiple brains(“hyperscanning”) have also been developed,which can be used to study multiple-brain-level differences in activity in games andmarkets (Read Montague et al. 2002).

2.2 Single-Neuron Measurement

Even the finest-grained brain imagingtechniques only measure activity of “cir-cuits” consisting of thousands of neurons. In



single neuron measurement, tiny electrodesare inserted into the brain, each measuring asingle neuron’s firing. As we discuss below,single neuron measurement studies haveproduced some striking findings that, webelieve, are relevant to economics. A limita-tion of single neuron measurement is that,because insertion of the wires damages neurons, it is largely restricted to animals.

Studying animals is informative abouthumans because many brain structures andfunctions of non-human mammals are simi-lar to those of humans (e.g., we are moregenetically similar to many species of mon-keys than those species are to other species).Neuroscientists commonly divide the braininto crude regions that reflect a combinationof evolutionary development, functions, andphysiology. The most common, triune divi-sion draws a distinction between the “reptil-ian brain,” which is responsible for basicsurvival functions, such as breathing, sleep-ing, eating, the “mammalian brain,” whichencompasses neural units associated withsocial emotions, and the “hominid” brain,which is unique to humans and includesmuch of our oversized cortex—the thin, fold-ed, layer covering the brain that is responsi-ble for such “higher” functions as language,consciousness and long-term planning (PaulMacLean 1990). Because single neuronmeasurement is largely restricted to nonhu-man animals, it has so far shed far more lighton the basic emotional and motivationalprocesses that humans share with othermammals than on higher-level processessuch as language and consciousness.

2.3 Electrical Brain Stimulation (EBS)

Electrical brain stimulation is anothermethod that is largely restricted to animals.In 1954, two psychologists (James Olds andPeter Milner 1954) discovered that ratswould learn and execute novel behaviors ifrewarded by brief pulses of electrical brainstimulation (EBS) to certain sites in thebrain. Rats (and many other vertebrates,including humans) will work hard for EBS.

For a big series of EBS pulses, rats will leapover hurdles, cross electrified grids, andforego their only daily opportunities to eat,drink, or mate. Animals also trade EBS offagainst smaller rewards in a sensible fash-ion—e.g., they demand more EBS to foregofood when they are hungry. Unlike morenaturalistic rewards, EBS does not satiate.And electrical brain stimulation at specificsites often elicits behaviors such as eating,drinking (Joseph Mendelson 1967), or copu-lation (Anthony Caggiula and BartleyHoebel 1966). Many abused drugs, such ascocaine, amphetamine, heroin, cannabis,and nicotine, lower the threshold at whichanimals will lever-press for EBS (Roy Wise1996). Despite its obvious applications toeconomics, only a few studies have exploredsubstitutability of EBS and other reinforcers(Steven Hursh and B. H. Natelson; LeonardGreen and Howard Rachlin 1991; PeterShizgal 1999).

2.4 Psychopathology and Brain Damage inHumans

Chronic mental illnesses (e.g., schizophre-nia), developmental disorders (e.g., autism),degenerative diseases of the nervous system,and accidents and strokes that damage local-ized brain regions help us understand howthe brain works (e.g., Antonio Damasio1994). When patients with known damage toan area X perform a special task more poor-ly than “normal” patients, and do other tasksequally well, one can infer that area X is usedto do the special task. Patients who haveundergone neurosurgical procedures such aslobotomy (used in the past to treat depres-sion) or radical bisection of the brain (anextreme remedy for epilepsy, now rarelyused) have also provided valuable data (seeWalter Freeman and James Watts 1942;Gazzaniga and LeDoux 1978).

Finally, a relatively new method calledtranscranial magnetic stimulation (TMS)uses pulsed magnetic fields to temporarilydisrupt brain function in specific regions.The difference in cognitive and behavioral



functioning that results from such disrup-tions provides clues about which regionscontrol which neural functions. The theoret-ical advantage of TMS over brain imaging isthat TMS directly leads to causal inferencesabout brain functioning rather than thepurely associational evidence provided byimaging techniques. Unfortunately, the useof TMS is currently limited to the cortex (itis particularly useful for studying visualprocesses in the occipital lobe, in the back ofthe head). It is also controversial because itcan cause seizures and may have other badlong-run effects.

2.5 Psychophysical Measurement

An old and simple technique is measure-ment of psychophysiological indicators likeheart rate, blood pressure, galvanic skinresponse (GSR, sweating in the palms), andpupil dilation (pupils dilate in response toarousal, including monetary reward). Thesemeasurements are easy, not too obtrusive,portable, and very rapid in time. The draw-back is that measurements can fluctuate formany reasons (e.g., body movement) andmany different combinations of emotionslead to similar psychophysiological output,just as being pulled over by a cop and meet-ing a blind date may produce very similaremotional anxiety responses. Often thesemeasurements are useful in combinationwith other techniques or in lesion patientswho are likely to have very different physio-logical reactions (e.g., sociopaths do notshow normal GSR fear reactions before apossible monetary loss). Facial musculaturecan also be measured by attaching smallelectrodes to smiling muscles (on the cheekbones) and frowning muscles (betweenthe eyebrows).

2.6 Diffusion Tensor Imaging (DTI)

Diffusion tensor imaging (DTI) is a newtechnique which exploits the fact that waterflows rapidly though myleinated (sheathed)neural axons (Dennis Bihan et al. 2001).Imaging the water flow can therefore reveal

the trajectories which project from one neu-ral region to others (like watching patternsof car traffic from a helicopter can tell youroughly the ebb and flow of economic andsocial activity). Knowing where a region’sneurons project is extremely useful inunderstanding neural circuitry and is animportant complement to simply imagingactivity in multiple regions (using fMRI)with little ability to pin down which activityoccurs earliest. Furthermore, the techniquecan be used after autopsies, which is anobvious advantage.

2.7 Is Neuroscience Just About WhereThings Happen in the Brain?

Neuroscience is sometimes criticized asproviding little more than a picture of“where things happen in the brain” or, morecynically, as simply showing that behavior iscaused by action in the nervous system(which was never in doubt). Indeed, someneuroscientists who are purely interested inthe functionality of different brain regionswould endorse such a characterizationunapologetically.

However, the long-run goal of neuro-science is to provide more than a map of themind. By tracking what parts of the brainare activated by different tasks, and espe-cially by looking for overlap betweendiverse tasks, neuroscientists are gaining anunderstanding of what different parts of thebrain do, how the parts interact in “circuit-ry,” and, hence, how the brain solves differ-ent types of problems. For example,because different parts of the brain aremore or less associated with affective orcognitive processing (a distinction wedefine more precisely below), imaging peo-ple while they are doing different types ofeconomic tasks provides important cluesabout the mix of affective and cognitiveprocesses in those tasks. For example,Sanfey et al. (2003) find that the insula cor-tex—a region in the temporal lobe thatencodes bodily sensations like pain andodor disgust—is active when people receive



low offers in an ultimatum bargaininggame. This means that even if rejecting alow offer is done because of an adaptedinstinct to build up a reputation for tough-ness (in order to get more in the future),the circuitry that encodes this instinct clear-ly has an affective component, which is notpurely cognitive.

For neuroeconomists, knowing moreabout functional specialization, and howregions collaborate in different tasks, couldsubstitute familiar distinctions between cat-egories of economic behavior (sometimesestablished arbitrarily by suggestions whichbecome modeling conventions) with newones grounded in neural detail. For exam-ple, the insula activity noted by Sanfey et al.in bargaining is also present when subjectsplaying matrix games are asked to guesswhat other subjects think they will do (sec-ond-order beliefs) (see Meghana Bhatt andCamerer, in press). This suggests that maybethe subjects who receive low offers in theultimatum study are not disgusted, they aresimply evaluating a second-order beliefabout what proposers expect them to do, asan input into an emotional evaluation. This isjust a speculation, of course, but it showshow direct understanding of neural circuitrycan inspire theorizing and the search fornew data.

In light of the long list of methodsreviewed earlier in this section, it is alsoworth emphasizing that neuroscience isn’tonly about brain imaging. Brain lesion stud-ies (as well as TMS) allow one to examinethe impact of disabling specific parts of thebrain. They have often provided clearer evi-dence on functionality of specific brainregions—even with tiny sample sizes fromrare types of damage—than imaging meth-ods have. Single neuron measurement pro-vides information not just about what partsof the brain “light up,” but about the specif-ic conditions that cause specific neurons tofire at accelerated or decelerated rates. Andelectrical brain stimulation, though it islargely off limits to studies involving

humans, provides a level of experimentalcontrol that economists conducting fieldresearch should envy.

As always, “the proof is in the pudding.”Here, our goal is not to review the manyways in which neuroscience will rapidlychange economic theory, because that is notwhere we are yet. Our goal is to showcasesome key findings in neuroscience, and stim-ulate the reader’s curiosity about what thesefindings might mean for economics.

3. Basic Lessons from Neuroscience

Because most of these techniques involvelocalization of brain activity, this can easilyfoster a misperception that neuroscience ismerely developing a “geography of thebrain,” a map of which brain bits do whatpart of the job. If that were indeed so, thenthere would be little reason for economiststo pay attention. In reality, however, neuro-science is beginning to use regional activitydifferences and other clues to elucidate theprinciples of brain organization and func-tioning, which in turn, is radically changingour understanding of how the brain works.Our goal in this section is to highlight someof the findings from neuroscience that mayprove relevant to economics.

3.1 A Two-Dimensional TheoreticalFramework

Our organizing theme, depicted in Table 1,emphasizes the two distinctions mentionedin the introduction, between controlled andautomatic processes, and between cognitionand affect.

3.1.1 Automatic and Controlled Processes

The distinction between automatic andcontrolled processes was first proposed bySchneider and Shiffrin (1977). Many othershave developed similar two-system modelssince then, with different labels: rule-basedand associative (Steven Sloman 1996);rational and experiential systems (LeeKirkpatrick and Seymour Epstein 1992);



TABLE 1TWO DIMENSIONS OF NEURAL FUNCTIONING

Cognitive Affective

Controlled Processes■ serial■ effortful■ evoked deliberately I II■ good introspective

access

Automatic Processes■ parallel■ effortless■ reflexive III IV■ no introspective

access

2 Elaborate methods have been developed that maxi-mize the validity of such “verbal protocols” (see, e.g.,Herbert Simon).

reflective and reflexive (MatthewLieberman et al. 2002); deliberative andimplementive systems (Peter Gollwitzer,Kentaro Fujita, and Gabriele Oettingen2004); assessment and locomotion (ArieKruglanski et al. 2000), and type I and typeII processes (Daniel Kahneman and ShaneFrederick 2002).

Controlled processes, as described by thetwo rows of Table 1, are serial (they usestep-by-step logic or computations), tend tobe invoked deliberately by the agent whenher or she encounters a challenge or sur-prise (Reid Hastie 1984), and are oftenassociated with a subjective feeling ofeffort. People can typically provide a goodintrospective account of controlled process-es. Thus, if asked how they solved a mathproblem or choose a new car, they canoften recall the considerations and thesteps leading up to the choice.2 Standardtools of economics, such as decision treesand dynamic programming, can be viewedas stylized representations of controlledprocesses.

3 The brain’s ability to recover from environmentaldamage is also facilitated by a property called plasticity. Inone study that illustrates the power of plasticity, the opticnerves of ferrets (which are born when they are still at arelatively immature state of development when the brain isstill highly plastic) were disconnected at birth and recon-nected to the auditory cortex (the portion of the brain thatprocesses sound). The ferrets learned to “see” using audi-tory cortex, and some neurons in their auditory cortexactually took on the physical characteristics of neurons inthe visual cortex (Lauire von Melchner, Sarah Pallas, andMriganka Sur 2000).

Automatic processes are the opposite ofcontrolled processes on each of thesedimensions— they operate in parallel, arenot accessible to consciousness, and arerelatively effortless. Parallelism facilitatesrapid response, allows for massive multi-tasking, and gives the brain remarkablepower when it comes to certain types oftasks, such as visual identification.Parallelism also provides redundancy thatdecreases the brain’s vulnerability to injury.When neurons are progressively destroyedin a region, the consequences are typicallygradual rather than sudden (“gracefuldegradation”).3 “Connectionist” neural net-work models formulated by cognitive psy-chologists (David Rumelhart and JamesMcClelland 1986) capture these featuresand have been applied to many domains,



Figure 1. The human brain with some economically relevant areas marked.

including commercial ones. Models of thistype have a very different structure fromthe systems of equations that economiststypically work with. Unlike systems ofequations, they are “black-box”; it is hard tointuit what they are doing by looking atindividual parameters.

Because automatic processes are notaccessible to consciousness, people oftenhave surprisingly little introspective insightinto why automatic choices or judgmentswere made. A face is perceived as “attrac-tive” or a verbal remark as “sarcastic” auto-matically and effortlessly. It is only laterthat the controlled system may reflect onthe judgment and attempt to substantiate itlogically, and when it does, it often does sospuriously (e.g., Timothy Wilson, SamuelLindsey, and Tonya Schooler 2000).

Automatic and controlled processes canbe roughly distinguished by where theyoccur in the brain (Lieberman et al. 2002).Regions that support cognitive automaticactivity are concentrated in the back (occip-ital), top (parietal), and side (temporal)parts of the brain (see Figure 1). The amyg-dala, buried below the cortex, is responsiblefor many important automatic affectiveresponses, especially fear. Controlledprocesses occur mainly in the front (orbitaland prefrontal) parts of the brain. The pre-frontal cortex (pFC) is sometimes called the“executive” region, because it draws inputsfrom almost all other regions, integratesthem to form near and long-term goals, andplans actions that take these goals intoaccount (Timothy Shallice and PaulBurgess 1996). The prefrontal area is the

Somato-sensory

VisualcortexAutomatic

affect

Cognitivecontrol“interrupt

Brodmann”10

L insula R insula



4 As David Laibson and Andrew Caplin, respectively,have aptly expressed it.

region that has grown the most in thecourse of human evolution and which,therefore, most sharply differentiates usfrom our closest primate relatives (StephenManuck et al. 2003).

Automatic processes—whether cognitiveor affective—are the default mode of brainoperation. They whir along all the time,even when we dream, constituting most ofthe electro-chemical activity in the brain.Controlled processes occur at specialmoments when automatic processes become“interrupted,” which happens when a per-son encounters unexpected events, experi-ences strong visceral states, or is presentedwith some kind of explicit challenge in theform of a novel decision or other type ofproblem. To the degree that controlledprocesses are well described by economiccalculation but parallel processes are not,one could say that economics is about the“interrupt” or “override.” 4

3.1.2 Affective and Cognitive Processes

The second distinction, represented bythe two columns of table 1, is betweenaffective and cognitive processes. Such adistinction is pervasive in contemporarypsychology (e.g., Robert Zajonc 1980, 1984,1998; Zajonc and Daniel McIntosh 1992)and neuroscience (Damasio 1994; LeDoux1996; Panksepp 1998), and has an historicallineage going back to the ancient Greeksand earlier (Plato described people as driv-ing a chariot drawn by two horses, reasonand passions).

The distinguishing features of affectiveprocessing are somewhat counterintuitive.Most people undoubtedly associate affectwith feeling states, and indeed most affectstates do produce feeling states when theyreach a threshold level of intensity.However, most affect probably operatesbelow the threshold of conscious awareness

5 By this definition, neural processes that don’t havevalence are not affects. There are also neural processesthat produce actions that are not best defined as affects –e.g., the reflexes that cause you to draw away from a hotobject or an electric shock.

(LeDoux 1996; Piotr Winkielman and KentBerridge 2004). As Rita Carter (1999) com-ments, “the conscious appreciation of emo-tion is looking more and more like onequite small, and sometimes inessential, ele-ment of a system of survival mechanismsthat mainly operate—even in adults—at anunconscious level.”

For most affect researchers, the centralfeature of affect is not the feeling states asso-ciated with it, but its role in human motiva-tion. All affects have “valence”—they areeither positive or negative (though somecomplex emotions, such as “bittersweet,”can combine more basic emotions that haveopposing valences). Many also carry “actiontendencies” (Nico Frijda 1986; LeonardBerkowitz 1999)—e.g., anger motivates us toaggress, pain to take steps to ease the pain,and fear to escape or in some cases tofreeze—as well as diverse other effects onsensory perception, memory, preferences,and so on (see, e.g., Leda Cosmides andJohn Tooby 2004). Affective processes,according to Zajonc’s (1998) definition, arethose that address “go/no-go” questions—that motivate approach or avoidance behav-ior. Cognitive processes, in contrast, arethose that answer true/false questions.5

Though it is not essential to our overall argu-ment, our view is that cognition by itself can-not produce action; to influence behavior,the cognitive system must operate via theaffective system.

Affect, as we use the term, embodies notonly emotions such as anger, fear, and jeal-ousy, but also drive states such as hunger,thirst and sexual desire, and motivationalstates such as physical pain, discomfort(e.g., nausea) and drug craving. Ross Buck(1999) refers to these latter influences as“biological affects,” which he distinguishes



6 The researchers scanned the brains of subjects usingfMRI (functional magnetic resonance imaging) as theyplayed a video game designed to produce a feeling of socialrejection. Subjects thought they were playing a game thatinvolved throwing a ball back and forth with two otherpeople, but in fact the computer controlled the two ani-mated figures that they saw on the screen. After a periodof three-way play, the two other “players” began to excludethe subjects by throwing the ball back and forth betweenthemselves. The social snub triggered neural activity in apart of the brain called the anterior cingulate cortex, whichalso processes physical pain, and the insula, which is activeduring physical and social discomfort.

from the more traditional “social affects.”Affect, thus, coincides closely with the his-torical concept of the passions. Althoughemotions such as anger and fear mightseem qualitatively different than the bio-logical affects, they have more in commonthat might be supposed. Thus, a recentstudy showed that hurt feelings activatedthe same brain regions activated by brokenbones or other physical injuries (NaomiEisenberger et al. 2003).6

3.1.3 The Quadrants in Action: AnIllustration

These two dimensions, in combination,define four quadrants (labeled I to IV inTable 1). Quadrant I is in charge when youdeliberate about whether to refinance yourhouse, poring over present-value calcula-tions; quadrant II is undoubtedly the rarestin pure form. It is used by “method actors”who imagine previous emotional experi-ences so as to actually experience thoseemotions during a performance; quadrantIII governs the movement of your hand asyou return serve; and quadrant IV makesyou jump when somebody says “Boo!”

Most behavior results from the interac-tion of all four quadrants. A natural instinctof economists trained in parsimoniousmodeling is to think that cognition is typi-cally controlled, and affect is automatic, sothere are really only two dimensions(quadrants I and IV) rather than four. Buta lot of cognitive processing is automatic aswell—e.g., visual perception or language

7 For example, when people are shown unpleasant pho-tographs and told to interpret the photos so that they don’texperience negative emotions (e.g., imagine a picture ofwomen crying as having been taken at a wedding), there isless activity in the insula and amygdala, emotional areas,and in medial orbitofrontal cortex, and area which theinsula projects to (Kevin Ochsner et al., in press).

processing. Research on “emotional regu-lation” shows many ways that cognitioninfluences emotion, which implies thecapacity for controlling emotion.7

The four-quadrant model is just a way toremind readers that the cognitive–affectiveand controlled–automatic dimensions arenot perfectly correlated, and to provide abroad view to guide exploratory research.For some purposes, reducing the twodimensions to one, or the four quadrants totwo, will certainly be useful. Furthermore,noting all four cells is not a claim that all areequally important. It is just a suggestion thatleaving out one of the combinations wouldlead to a model which is incomplete forsome purposes.

Consider what happens when a party hostapproaches you with a plate of sushi.

Quadrant III: Your first task is to figureout what is on the plate. The occipital cor-tex in the back of the brain is the first on thescene, drawing in signals from your eyes viayour optic nerves. It decodes the sushi intoprimitive patterns such as lines and cornersthen uses a “cascading process” to discernlarger shapes (Stephen Kosslyn 1994).Further downstream, in the inferior tempo-ral visual cortex (ITVC), this informationbecomes integrated with stored representa-tions of objects, which permits you to rec-ognize the objects on the plate as sushi. Thislatter process is extraordinarily complicated(and has proved difficult for artificial intelli-gence researchers to recreate in computers)because objects can take so many forms,orientations, and sizes.

Quadrant IV: This is where affect entersthe picture. Neurons in the inferior tempo-ral visual cortex are sensitive only to the



identity of an object; they don’t tell youwhether it will taste good. Outputs of theinferior temporal visual cortex as well as out-puts from other sensory systems feed intothe orbitofrontal cortex to determine the“reward value of the recognized object.”This is a highly particular representation. Ineconomic terms, what is represented is nei-ther pure information (i.e., that this is sushi)nor pure utility (i.e., that it is something Ilike) but rather a fusion of information andutility. It is as if certain neurons in theorbitofrontal cortex are saying “this is sushiand I want it.”

The reward value of sushi depends inturn on many factors. First, there is yourpersonal history with sushi. If you got sickon sushi in the past, you will have an uncon-scious and automatic aversion to it (“tasteaversion conditioning”). The amygdalaseems to play a critical role in this kind oflong-term learning (LeDoux 1996).Second, the reward value of the sushi willdepend on your current level of hunger;people can eat almost anything—grass,bugs, human flesh—if they are hungryenough. The orbitofrontal cortex and a sub-cortical region called the hypothalamus aresensitive to your level of hunger (Rolls1999). Neurons in these regions fire morerapidly at the sight or taste of food whenyou are hungry, and fire less rapidly whenyou are not hungry.

Quadrants I and II: Processing oftenends before quadrants I and II go to work.If you are hungry, and like sushi, yourmotor cortex will guide your arm to reachfor the sushi and eat it, drawing on auto-matic quadrant III (reaching) and IV(taste and enjoyment) processes. Undersome circumstances, however, higher levelprocessing may enter the picture. If yousaw a recent documentary on the risks ofeating raw fish, you may recoil; or if youdislike sushi but anticipate disappointmentin the eyes of your proud host who madethe sushi herself, you’ll eat it anyway (orpick it up and discreetly hide it in a napkin

8 Paul Romer (2000) uses the example of peanut tastesas an illustration of how understanding the cause ofrevealed preference matters. One person loves the taste ofpeanuts, but is allergic to them and knows that the conse-quences of eating would be disastrous. When she is hun-gry, her visceral system motivates her to eat peanuts, buther deliberative system, with its ability to consider delayedconsequences, inhibits her from eating them. The otherperson developed a “taste aversion” to peanuts many yearsago, as a result of having gotten sick right after eatingthem. She knows at a cognitive level that the peanuts werenot the cause of her sickness, but her visceral system over-rules cognitive awareness. Revealed preference theorywould stop at the conclusion that both women have disu-tility for peanuts. But the fact that the mechanisms under-lying their preferences are different leads to predictabledifferences in other kinds of behavior. For example, thetaste-averse woman will have a higher price elasticity(she’ll eat peanuts if paid enough) and she will learn toenjoy peanuts after eating them a few times (her taste-aversion can be “extinguished”). The allergic woman willalso like the smell of peanuts, which the taste-averterwon’t. Treatments also differ: cognitive therapies for treat-ing harmless phobias and taste-aversion train people to usetheir conscious quadrant I processing to overrule visceralquadrant IV impulses, whereas treatment of the allergicpeanut-avoider will concentrate on a medical cure.

when she turns to serve other guests).8

These explicit thoughts involve anticipatedfeelings (your own and the host’s) anddraw on explicit memories from a part ofthe brain called the hippocampus (see fig-ure 1), inputs from the affective system(sometimes referred to as the “limbic sys-tem”), and anticipation (planning) fromthe prefrontal cortex.

Because standard economics is bestdescribed by the controlled, cognitive,processes of quadrant 1, in the remainder ofthis section we focus on the other half ofeach dichotomy—automatic and affectiveprocesses—providing further details of theirfunctioning.

3.2 Automatic Processes

Here, we review some key principles ofneural functioning that characterize auto-matic processes. Our short-list includes:parallelism, specialization, and coordina-tion. Unpacking this a bit, we would say that:(1) much of the brain’s processing involvesprocesses that unfold in parallel and are not



accessible to consciousness; (2) the brainutilizes multiple systems specialized to per-form specific functions; and (3) it figures outhow to use existing specialized systems toaccomplish new tasks efficiently, whateverfunctions they originally evolved to perform.

3.2.1 Parallelism

The brain performs a huge number of dif-ferent computations in parallel. Because of themassively interconnected “network” architec-ture of neural systems, computations done inone part of the brain have the potential toinfluence any other computation, even whenthere is no logical connection between the two.

Recent psychological research on automat-ic processing provides many striking examplesof such spurious interactions. In one particu-larly clever study, Gary Wells and RichardPetty (1980) had subjects listen to an editori-al delivered over headphones while shakingtheir head either up–down, or right–left (sub-jects were led to believe that shaking was alegitimate part of the “product test”). Thosewho were instructed to shake up–downreported agreeing with the editorial morethan those instructed to shake left–right, pre-sumably because in our culture up–downhead movement is associated with an attitudeof acceptance, and left–right with an attitudeof rejection. A similar impact on preferencewas also observed in a study in which subjectswere asked to evaluate cartoons while holdinga pen either clamped horizontally betweentheir teeth or grasping it with pursed lips as ifpuffing on a cigarette (Fritz Strack, LeonardMartin, and Sabine Stepper 1988). The hori-zontal clamp forces the mouth into a smile,which enhances ratings, while “puffing”forces the mouth into a pursed expression,which lowers ratings. What the brain seems tobe doing, in these examples, is seeking a“global equilibrium” that would reconcile theforced action (e.g., forced smile) with theresponse and the perceived attributes of theevaluated object.

Given that people are capable of delibera-tion, why doesn’t quadrant I thinking correct

the automatic activity produced by quadrantIII and IV processes when they are wrong?Indeed sometimes it does: pilots learn totrust their instrument panels, even whenthey conflict with strong sensory intuitionsabout where their plane is headed, but suchcognitive override is more the exceptionthan the rule. To override automaticprocesses, quadrant I has to (a) recognizethat an initial impression is wrong (whichrequires self-awareness about behavior inthe other quadrants), and then (b) deliber-ately correct that impression. But whensense making works outside of consciousnessit will not generate alarm bells to trigger therecognition required in (a). This is surely thecase in the Epley and Gilovich studies. Evenwhen the external influence is obvious andinappropriate, or the subject is warnedahead of time, the deliberation required tocorrect the first impression is hard work,competing for mental resources and atten-tion with all the other work that needs to bedone at the same moment (Daniel Gilbert2002). The struggle between rapid uncon-scious pattern-detection processes and theirslow, effortful modulation by deliberation isnot a fair contest; so automatic impressionswill influence behavior much of the time.

3.2.2 Specialization

Neurons in different parts of the brain havedifferent shapes and structures, different func-tional properties, and operate in coordinationas systems that are functionally specialized.Progress in neuroscience often involves trac-ing well-known psychological functions to cir-cumscribed brain areas. For example, Broca’sand Wernicke’s areas are involved, respective-ly, in the production and comprehension oflanguage. Patients with Wernicke damage can-not understand spoken words. While they canproduce words themselves, they can’t monitortheir own speech, which results in sentencesof correctly articulated words strung togetherinto unintelligible gibberish. People with dam-age to Broca’s area, in contrast, can understandwhat is said to them and typically know what



9 Retrograde amnesia is the more familiar kind, inwhich people forget their past but can form new memo-ries. The 2000 movie “Memento” drew an accurate por-trait of anterograde amnesia and—interestingly foreconomists— its vulnerability to exploitation by peoplewho know you are forgetting what they did to you.

they want to say, but have difficulty articulat-ing it, sometimes to the point of not being ableto generate any words at all.

Beyond uncovering the nature of thesespecialized systems, neuroscience has led tothe discovery of new functional systems,some of which are quite surprising. Forexample, surgeons conducting brain surgeryon an epileptic patient discovered a smallregion of her brain which, when stimulated,caused her to laugh (Itzhak Fried 1998),hinting at the existence of a “humor system.”Neuroscientists have also located an area inthe temporal lobe that, when stimulatedelectrically, produces intense religious feel-ings—e.g., the sense of a holy presence oreven explicit visions of God or Christ, evenin otherwise unreligious people (MichaelPersinger and Faye Healey 2002).

More generally, neuroscience has begunto change our classification of functionalprocesses in the brain, in some cases identi-fying distinct brain processes that serve thesame function, and in others drawing con-nections between processes that have beencommonly viewed as distinct.

An example of the former is memory.Studies of patients with localized brain dam-age have confirmed the existence of distinctmemory systems, which can be selectively“knocked out.” Anterograde amnesiacs, forexample, are commonly able to recall infor-mation acquired before injury, and they areable to acquire implicit information, includ-ing perceptual-motor skills (like the ability toread text in a mirror), but they are unable torecall new explicit information for more thana minute.9 Anterograde amnesiacs can alsoacquire new emotional associations withoutthe explicit memories necessary to makesense of them. In one famous example, a doc-tor introduced himself with a tack concealed

in his hand that pricked a patient when theyshook hands. Later the doctor returned and,although he did not remember the doctor, henevertheless responded negatively to thedoctor’s arrival and refused to shake his hand.

An example of the latter involves the expe-rience of fear, and recognition of fear in oth-ers. Experiencing an emotion andrecognizing the emotion when displayedintuitively seems to involve entirely differentprocesses. However, recent findings suggestthat this is not the case. Numerous studieshave implicated the amygdala—a small“organ” in the brain that is also intimatelylinked to the sense of smell—to fear pro-cessing. Lesions to the amygdala of rats andother animals disrupt or even eliminate theanimals’ fear-responses. Humans with strokedamage in the amygdala show similardeficits in reacting to threatening stimuli.

The same damage that disrupts fearresponses in humans also cripples a person’sability to recognize facial expressions of fearin others, and to represent such expressionsin pictures. Figure 2 shows how a patientwith amygdala damage expressed differentemotions when asked to draw them in theform of facial expressions. The patient wasable to render most emotions with remark-able artistic talent. But when it came todrawing an expression of fear she felt clue-less. She didn’t even try to draw an adultface; instead she drew a picture of a crawlinginfant looking apprehensive.

The finding that people who don’t experi-ence fear also can’t recognize it or represent itpictorially, suggests that two phenomena thathad been viewed as distinct—experiencingand representing fear—in fact have importantcommonalities. Beyond this, they point to theintriguing notion that to recognize an emotionin others, one needs to be able to experienceit oneself (see Alvin Goldman 2003).

The idea that people have specialized sys-tems that are invoked in specific situationscould have dramatic consequences for eco-nomics. The standard model of economicbehavior assumes that people have a unitary



Figure 2. Representations of different emotions drawn by a patient with amygdala damage (Adolphs et al. 1995).

set of preferences which they seek to satisfy,and economists often criticize psychology forlacking such a unified perspective. The exis-tence of such selectively—invoked, special-ized, systems, however, raises the questionof whether a unified account of behavior is likely to do a very good job of capturingthe complexities of human behavior. AsJonathan Cohen (personal communication)aptly expressed it at a recent conference onneuroeconomics, “Economics has one theo-ry, psychologists many—perhaps because

brains have different systems that they use tosolve different problems.”

3.2.3 Coordination

In a process that is not well understood,the brain figures out how to do the tasks it isassigned efficiently, using the specialized sys-tems it has at its disposal. When the brain isconfronted with a new problem it initiallydraws heavily on diverse regions, including,often, the prefrontal cortex (where controlled



Figure 3. Regions of brain activation when first playing Tetris (left) and after several weeks of practice (right) (Haier et al. 1992).

processes are concentrated). But over time,activity becomes more streamlined, concen-trating in regions that specialized in process-ing relevant to the task. In one study(Richard Haier et al. 1992), subjects’ brainswere imaged at different points in time asthey gained experience with the computergame Tetris, which requires rapid hand–eyecoordination and spatial reasoning. Whensubjects began playing, they were highlyaroused and many parts of the brain wereactive (Figure 3, left panel). However, as theygot better at the game, overall blood flow tothe brain decreased markedly, and activitybecame localized in only a few brain regions(Figure 3, right panel).

Much as an economy ideally adjusts to theintroduction of a new product by graduallyshifting production to the firms that can pro-duce the best goods most cheaply, with expe-rience at a task or problem, the brain seemsto gradually shift processing toward brainregions and specialized systems that cansolve problems automatically and efficientlywith low effort.

Andrew Lo and Dmitry Repin (2002)observe a similar result in a remarkablestudy of professional foreign-exchange andderivatives traders who were wired for psy-chophysiological measurements while theytraded. Less-experienced traders showedsignificant physiological reactions to abouthalf of the market events (e.g., trend rever-sals). More experienced traders reactedmuch less to the same events. Years ofautomatizing apparently enabled seasonaltraders to react calmly to dramatic eventsthat send a novice trader on an emotionalroller coaster.

Given the severe limitations of controlledprocesses, the brain is constantly in theprocess of automating the processing oftasks—i.e., executing them using automaticrather than controlled processes. Indeed,one of the hallmarks of expertise in an area isthe use of automatic processes such as visualimagery and categorization. In one now-famous study, Fernand Gobet and Simon(1996) tested memory for configurations ofchess pieces positioned on a chessboard.



They found that expert chess players wereable to store the positions of players almostinstantly—but only if they were in positionscorresponding to a plausible game. For ran-domly arranged chess pieces, the expertswere not much better than novices. Furtherinvestigations have found that chess grand-masters store roughly 10,000 different possi-ble board setups in memory which they canrecognize almost instantly and respond to.More recent research in decision makingsuggests that this is a more general phenom-enon, because decision making often takesthe form of pattern matching rather than ofan explicit weighing of costs and benefits(e.g., Robyn Leboeuf 2002; Doug Medinand Max Bazerman 1999).

In some cases, repeated use of particularspecialized systems can produce physicallyrecognizable changes. Studies have foundthat, for example, violinists who finger violinstrings with their left hand show enlargeddevelopment of cortical regions which corre-spond to fingers on the left hand (ThomasElbert et al. 1995), and the brain regionsresponsible for navigation and spatial mem-ory (the hippocampus) of London taxi driv-ers are larger than comparable areas innon-taxi drivers (Eleanor Maguire et al.2000). While the exact causality is difficult todetermine (perhaps preexisting differencesin size of brain region affect people’s talentsand occupational choices), a causal connec-tion in the posited direction has beendemonstrated in songbirds, whose brainsshow observable differences as a function ofwhether or not they have been exposed tothe song that is characteristic of their species(Carol Whaling et al. 1997).

3.2.4 The Winner Take All Nature ofNeural Processing

Another feature that is reminiscent of theoperation of an economy is the “winner-take-all” nature of neural information processing(M. James Nichols and Bill Newsome 2002).While there is a lot that we do not knowabout the way in which the huge amount of

10 Alternatively, one could imagine that perceptions orbeliefs aggregate all relevant neural information (whichwould also be more in tune with Bayesian updating). So faras we can tell, the brain can use both principles of aggre-gation: when different neural populations carry “similar”information, then the overall perceptual judgment is anaverage of all information; when they carry very differentinformation, then the overall judgment follows the “win-ner-take-all” rule (Nichols and Newsome 2002). For a sim-ple model of a neural network that exhibits theseproperties, see Richard Hahnloser et al. 2000.

neural activity is distilled into a categoricalpercept, or a decision, we do know that thebrain doesn’t invariably integrate (i.e., aver-age) over the signals carried by individualneurons. In particular, when two distinctgroups of neurons convey different informa-tion about the external world, the resultingperceptual judgment often adopts the infor-mation of one neuronal group and entirelysuppresses the information carried by theother. This is referred to as a “winner-take-all” principle of neural signal-extraction. As aresult, many brain processes are fundamen-tally categorical, yielding well-defined per-ceptions and thoughts even if the incominginformation is highly ambiguous.10 Theadvantages of this principle are clear if theultimate job of the brain is to initiate a dis-crete action, or categorize an object as one ofseveral discrete types. The disadvantage isthat updating of beliefs in response to newinformation can procede in fits and starts,with beliefs remaining static as long as thenew information does not lead to recatego-rization, but changing abruptly and dramati-cally when the accumulation of evidenceresults in a change of categorization (seeSendhil Mullainathan 2002).

3.3 Affective Processes

The way the brain evolved is critical tounderstanding human behavior. In manydomains, such as eating, drinking, sex, andeven drug use, human behavior resemblesthat of our close mammalian relatives, whichis not surprising because we share many ofthe neural mechanisms that are largelyresponsible for these behaviors. Many of the



processes that occur in these systems areaffective rather than cognitive; they aredirectly concerned with motivation. Thismight not matter for economics were it notfor the fact that the principles that guide theaffective system—the way that it operates—is so much at variance with the standard economic account of behavior.

3.3.1 Primacy of Affect

In contrast to the intuitive view of humanbehavior as driven by deliberations aboutcosts and benefits, it does not do a terribleinjustice to the field of psychology to say thata growing consensus has developed aroundthe view that affect is primary in the sensethat it is “first on the scene” and plays a dom-inant role in behavior. Indeed, as we discussbelow (section 3.4.2), the conscious brainoften erroneously interprets behavior thatemerges from automatic, affective, processesas the outcome of cognitive deliberations.

In a series of seminal papers, Zajonc (1980,1984, 1998) presented the results of studieswhich showed, first, that people can oftenidentify their affective reaction to some-thing—whether they like it or not—morerapidly than they can even say what it is, and,second, that affective reactions to things canbe dissociated from memory for details ofthose things, with the former often being bet-ter. For example, we often rememberwhether we liked or disliked a particular per-son, book or movie, without being able toremember any other details (Bargh 1984).Subsequent research in social psychologytakes Zajonc’s initial research a step furtherby showing that the human brain affectivelytags virtually all objects and concepts, andthat these affective tags are brought to mindeffortlessly and automatically when thoseobjects and concepts are evoked (e.g., RussellFazio et al. 1986; Anthony Greenwald, MarkKlinger, and Thomas Liu 1989; Greenwald1992; Bargh, Shelly Chaiken, PaulaRaymond, and Charles Hymes 1996; Jan DeHouwer, Dirk Hermans, and Paul Eelen1998; David Houston and Fazio 1989)

11 The amygdala has excellent properties for perform-ing such a function. Recent research in which amygdalaeof subjects were scanned while threatening stimuli wereflashed at various positions in the visual field found thatamygdala activation is just as rapid and just as pronouncedwhen such stimuli are presented outside as when they arepresented inside the region of conscious awareness (AdamAnderson et al. 2003).

LeDoux and his colleagues (summarized inLeDoux 1996) have arrived at similar conclu-sions about the primacy of affect using verydifferent research methods and subjects.Based on studies conducted on rats, they dis-covered that there are direct neural projec-tions from the sensory thalamus (whichperforms crude signal-processing) to theamygdala (which is widely believed to play acritical role in the processing of affectivestimuli) that are not channeled through theneocortex. As a result, animals can have anaffective reaction to stimuli before their cor-tex has had the chance to perform morerefined processing of the stimuli—they are lit-erally afraid before knowing whether theyshould be. Such immediate affective respons-es provide organisms with a fast but crudeassessment of the behavioral options they facewhich makes it possible to take rapid action.They also provide a mechanism for interrupt-ing and refocusing attention (Simon 1967),typically shifting processing from automatic tocontrolled processing. As Jorge Armony et al.(1995) note, “A threatening signal, such asone indicating the presence of a predator,arising from outside of the focus of attentionwill have a reduced representation in the cor-tex. Thus, if the amygdala relied solely on thecortical pathway to receive sensory informa-tion, it would not be capable of processing,and responding to, those danger signals thatare not within the focus of attention” (see also,Armony et al. 1997; Gavin DeBecker 1997).11

A similar pattern to the one LeDouxobserved in rats can also be seen in humans.Gilbert and Michael Gill (2000) propose thatpeople are “momentary realists” who trusttheir immediate emotional reactions andonly correct them through a comparativelylaborious cognitive process. If the car behind



12 The mechanisms that monitor the body’s state canbe exquisitely complex, and can include both internal andexternal cues. In the case of nutritional regulation, forexample, internal cues include gastric distension (JamesGibbs et al. 1981), receptors sensitive to the chemicalcomposition of the food draining from the stomach D.Greenberg, Smith and Gibbs (1990); Arthur L.Campfield and Smith 1990). External cues include timeof day, estimated time till the next feeding, and the sightor smell of food.

you honks after your light turns green, youare likely to respond with immediate anger,followed, perhaps, by a sheepish acknowl-edgment that maybe the honking personbehind you had a point, as you were dis-tracted when the light turned green.

It is important to note that while emotionsmay be fleeting, they can have a large eco-nomic impact if they create irreversible rashdecisions (as in “crimes of passion”). Evenemotions which could be transitory, such asembarrassment, can have long-run effects ifthey are kept alive by memory and socialreminders. For example, Dora Costa andMatthew Kahn (2004) found that desertersin the Union Army during the Civil War, whowere allowed to return to their hometownswithout explicit penalty, often moved awaybecause of the shame and social ostracism ofbeing known to their neighbors as a deserter.

3.3.2 Homeostasis

To understand how the affective systemoperates, one needs to recall that humans didnot evolve to be happy, but to survive andreproduce. An important process by whichthe body attempts to achieve these goals iscalled homeostasis. Homeostasis involvesdetectors that monitor when a system departsfrom a “set-point,”12 and mechanisms thatrestore equilibrium when such departuresare detected. Some—indeed most—of thesemechanisms do not involve deliberate action.Thus, when the core body temperature fallsbelow the 98.6° F set-point, blood tends tobe withdrawn from extremities, and when itrises above the set-point one begins to sweat.But other processes do involve deliberateaction—e.g., putting on one’s jacket when

cold or turning on the air conditioner whenhot. The brain motivates one to take suchactions using both a “carrot” and a “stick.”The stick reflects the fact that departingfrom a set-point usually feels bad—e.g., itfeels bad to be either excessively hot orcold—and this negative feeling motivatesone to take actions that move one backtoward the set-point. The carrot is a processcalled “alliesthesia” (Michael Cabanac1979) whereby actions that move onetoward the set-point tend to feel pleasura-ble. When the body temperature falls below98.6°, for example, almost anything thatraises body temperature (such as placingone’s hand in warm water) feels good, andconversely when the body temperature iselevated almost anything that lowers bodytemperature feels good.

As economists, we are used to thinking ofpreferences as the starting point for humanbehavior and behavior as the ending point. Aneuroscience perspective, in contrast, viewsexplicit behavior as only one of many mech-anisms that the brain uses to maintain home-ostasis, and preferences as transient statevariables that ensure survival and reproduc-tion. The traditional economic account ofbehavior, which assumes that humans act soas to maximally satisfy their preferences,starts in the middle (or perhaps even towardthe end) of the neuroscience account.Rather than viewing pleasure as the goal ofhuman behavior, a more realistic accountwould view pleasure as a homeostatic cue—an informational signal.13,14

13 Even deliberative behavior generated by quadrant Iis typically organized in a fashion that resembles home-ostasis (Miller, Eugene Gallanter, and Karl Pribram 1960;Loewenstein 1999). Rather than simply maximizing pref-erences in a limitless fashion, people set goals for them-selves, monitor their progress toward those goals, andadjust behavior when they fall short of their goals.

14 Of course, even the neuroscience account begins, insome sense, in the middle. A more complete understand-ing of behavior would also ask how these different mecha-nisms evolved over time. Since evolution selects for genesthat survive and reproduce, the result of evolution isunlikely to maximize pleasure or minimize pain. (See GaryBecker and Luis Rayo 2004.)



An important feature of many homeosta-tic systems is that they are highly attuned tochanges in stimuli rather than their levels. Adramatic demonstration of such sensitivityto change came from single-neuron studiesof monkeys responding to juice rewards(see Wolfram Schultz and AnthonyDickinson 2000). These studies measuredthe firing of dopamine neurons in the ani-mal’s ventral striatum, which is known toplay a powerful role in motivation andaction. In their paradigm, a tone was sound-ed, and two seconds later a juice reward wassquirted into the monkey’s mouth. Initially,the neurons did not fire until the juice wasdelivered. Once the animal learned that thetone forecasted the arrival of juice two sec-onds later, however, the same neurons firedat the sound of the tone, but did not firewhen the juice reward arrived. These neu-rons were not responding to reward, or itsabsence . . . they were responding to devia-tions from expectations. (They are some-times called “prediction neurons.”) Whenthe juice was expected from the tone, butwas not delivered, the neurons fired at avery low rate, as if expressing disappoint-ment. The same pattern can be observed ata behavioral level in animals, who will workharder (temporarily) when a reinforcementis suddenly increased and go “on strike”when reinforcement falls.15 Neural sensitiv-ity to change is probably important inexplaining why the evaluation of risky gam-bles depends on a reference point which

15 As Rolls (1999) writes, “We are sensitive to someextent not just to the absolute level of reinforcement beingreceived, but also to the change in the rate or magnitudeof reinforcers being received. This is well shown by thephenomena of positive and negative contrast effects withrewards. Positive contrast occurs when the magnitude of areward is increased. An animal will work very much hard-er for a period (perhaps lasting for minutes or longer) inthis situation, before gradually reverting to a rate close tothat at which the animal was working for the small rein-forcement. A comparable contrast effect is seen when thereward magnitude (or rate at which rewards are obtained)is reduced—there is a negative overshoot in the rate ofworking for a time.”

encodes whether an outcome is a gain or aloss (see section 5), why self-reported hap-piness (and behavioral indicators like sui-cide) depend on changes in income andwealth, rather than levels (Andrew Oswald1997), and why violations of expectationstrigger powerful emotional responses(George Mandler 1982).

3.4 Interactions between the Systems

Behavior emerges from a continuousinterplay between neural systems support-ing activity within each of the four quad-rants. Three aspects of this interaction bearspecial emphasis, which we labeled “collab-oration,” “competition,” and “sense-mak-ing.” Collaboration captures the insight thatdecision making, which is to say “rationality”in the broad, nontechnical sense of theword, is not a matter of shifting decision-making authority from quadrants II, III, andIV toward the deliberative, nonaffectivequadrant I, but more a matter of maintain-ing proper collaboration in activity across allfour quadrants. If quadrant I tries to do thejob alone, it will often fail.16 Competitionreflects the fact that different processes—most notably affective and cognitive—oftendrive behavior in conflicting directions andcompete for control of behavior. Sense-making refers to how we make sense of suchcollaboration and competition—how wemake sense of our behavior. While behavioris, in fact, determined by the interaction ofall four quadrants, conscious reflection onour behavior, and articulating reasons for it,is basically quadrant I trying to make senseof that interaction. And, not surprisingly,

16 Roy Baumeister, Todd Heatherton, and Dianne Tice(1994) review studies which suggest that excessively highincentives can result in supra-optimal levels of motivationthat have perverse effects on performance (known as theYerkes–Dodson law). Beyond documenting such “chokingunder pressure,” the review research showing that it oftenoccurs because it causes people to utilize controlledprocesses for performing functions, such as swinging a golfclub, that are best accomplished with automatic processes.See Dan Ariely et al. (2004) for the latest evidence.



quadrant I has a tendency to explain behav-ior in terms it can understand – in terms ofquadrant I processes.

3.4.1 Collaboration and Competition

Although it is heuristically useful to distin-guish between cognitive and affectiveprocesses, and between controlled and auto-matic processes, most judgments and behav-iors result from interactions between them.Collaboration, delegation of activity, andproper balance across the quadrants areessential to normal decision making. Manydecision-making disorders may originate inan improper division of labor between thequadrants. For example, psychiatry recog-nizes a decision-making continuum definedby the impulsive, “light” decision-makingstyle at one end and the compulsive, “heavy”style at the other. The decisions of an impul-sive individual are excessively influenced byexternal stimuli, pressures, and demands.Such a person may not be able to give a moresatisfying explanation of an action exceptthat “he felt like it” (David Shapiro 1965). Bycontrast, an obsessive–compulsive personwill subject even the most trivial decisions toextensive deliberation and calculation. In sit-uations in which it is entirely appropriate tomake a quick decision based “on impulse”—e.g., when choosing which video to rent for the evening or what to order in a restaurant—the obsessive-compulsive willget stuck.

We are only now beginning to appreciatethe importance of affect for normal decisionmaking. The affective system provides inputsin the form of affective evaluations of behav-ioral options—what Damasio (1994) refersto as “somatic markers.” Damasio and hiscolleagues show that individuals with mini-mal cognitive, but major affective deficitshave difficulty making decisions, and oftenmake poor decisions when they do (AntoineBechara et al. 1994; Bechara, Damasio,Damasio, and Lee 1999; Damasio 1994). It isnot enough to “know” what should be done;it is also necessary to “feel” it.

There is also interesting experimental evi-dence that deliberative thinking blocksaccess to one’s emotional reactions to objectsand so reduces the quality of decisions (e.g.,Wilson and Schooler 1991). In one study(Wilson et al. 1993) college students select-ed their favorite poster from a set of posters.Those who were instructed to think of rea-sons why they liked or disliked the postersbefore making their selection ended up lesshappy on average with their choice ofposters (and less likely to keep them up ontheir dorm room wall) than subjects whowere not asked to provide reasons.

Needless to say, affect can also distort cog-nitive judgments. For example, emotionshave powerful effects on memory—e.g.,when people become sad, they tend to recallsad memories (which often increases theirsadness). Emotions also affect perceptions ofrisks—anger makes people less threatenedby risks, and sadness makes them morethreatened (Jennifer Lerner and DacherKeltner 2001). Emotions also create “moti-vated cognition”—people are good at per-suading themselves that what they wouldlike to happen is what will happen. Quackremedies for desperate sick people, and get-rich-quick scams are undoubtedly aided bythe human propensity for wishful thinking.Wishful thinking may also explain high ratesof new business failure (Camerer and DanLovallo 1999), trading in financial markets,undersaving, and low rates of investment ineducation (foregoing large economicreturns). As LeDoux (1996) writes, “Whileconscious control over emotions is weak,emotions can flood consciousness. This is so because the wiring of the brain at thispoint in our evolutionary history is such thatconnections from the emotional systems tothe cognitive systems are stronger than con-nections from the cognitive systems to theemotional systems.”

When it comes to spending money ordelaying gratification, taking or avoidingrisks, and behaving kindly or nastily towardother people, people often find themselves



of “two minds”; our affective systems drive usin one direction and cognitive deliberationsin another. We find ourselves almost compul-sively eating our children’s left-overHalloween candy, while obsessing about howto lose the extra ten pounds; gambling reck-lessly at the casino, even as a small voice inour head tell us to rein it in; trying to build upcourage to step up to the podium; or tempt-ed to give a little to the pathetic street-cornerbeggar though we know our crumpled dollarwill go further if donated to United Way.

All of these deviations occur because ouraffective system responds to different cues,and differently to the same cues, as ourcognitive system does. As Rolls (1999)writes,

emotions often seem very intense in humans,indeed sometimes so intense that they producebehaviour which does not seem to be adaptive,such as fainting instead of producing an activeescape response, or freezing instead of avoiding,or vacillating endlessly about emotional situa-tions and decisions, or falling hopelessly in loveeven when it can be predicted to be withouthope or to bring ruin. The puzzle is not only thatthe emotion is so intense, but also that even withour rational, reasoning, capacities, humans stillfind themselves in these situations, and may findit difficult to produce reasonable and effectivebehaviour for resolving the situation (p. 282).

Such divergences between emotional reac-tions and cognitive evaluations arise, Rolls(1999) argues, because

in humans the reward and punishment systemsmay operate implicitly in comparable ways tothose in other animals. But in addition to this,humans have the explicit system (closely relatedto consciousness) which enables us consciouslyto look and predict many steps ahead. (p. 282).

Thus, for example, the sight or smell of acookie might initiate motivation to consumeon the part of the affective system, but mightalso remind the cognitive system that one ison a diet.

Exactly how cognitive and affective sys-tems interact in the control of behavior is not

well understood. At a neurological level, itappears that an organ in the reptilian braincalled the striatum (part of a larger systemcalled the basal ganglia) plays a critical role.The striatum receives inputs from all parts ofthe cerebral cortex, including the motor cor-tex, as well as from affective systems such asthe amygdala. Lesions of pathways that sup-ply dopamine to the striatum leads, in ani-mals, to a failure to orient to stimuli, a failureto initiate movements, and a failure to eat ordrink (J. F. Marshall et al. 1974). In humans,depletion of dopamine in the striatum isfound in Parkinson’s disease, the most dra-matic symptom of which is a lack of volun-tary movement. The striatum seems to beinvolved in the selection of behaviors fromcompetition between different cognitive andaffective systems—in producing one coher-ent stream of behavioral output, which canbe interrupted if a signal of higher priority isreceived, or a surprising stimulus appears(Carolyn Zink et al. 2003).

The extent of collaboration and competi-tion between cognitive and affective sys-tems, and the outcome of conflict when itoccurs, depends critically on the intensity ofaffect (Loewenstein 1996; Loewenstein andLerner 2003). At low levels of intensity,affect appears to play a largely “advisory”role. A number of theories posit that emo-tions carry information that people use as aninput into the decisions they face (e.g.,Damasio 1994; Ellen Peters and Paul Slovic2000). The best-developed of theseapproaches is affect-as-information theory(Norbert Schwarz and Gerald Clore 1983;Schwarz 1990; Clore 1992).

At intermediate levels of intensity, peo-ple begin to become conscious of conflictsbetween cognitive and affective inputs. Itis at such intermediate levels of intensitythat one observes the types of efforts atself-control that have received so muchattention in the literature (Jon Elster 1977;Walter Mischel, Ebbe Ebbesen, andAntonette Zeiss 1972; Thomas Schelling1978, 1984).



Finally, at even greater levels of intensity,affect can be so powerful as to virtually pre-clude decision making. No one “decides” tofall asleep at the wheel, but many people do.Under the influence of intense affectivemotivation, people often report themselves asbeing “out of control” or “acting against theirown self-interest” (Baumeister, Heatherton,and Tice 1994; Stephen Hoch andLoewenstein 1991; Loewenstein 1996). AsRita Carter writes, “where thought conflictswith emotion, the latter is designed by theneural circuitry in our brains to win” (1999).

3.4.2 Spurious Sense-Making

Sense-making is an important form ofinteraction between quadrant I and theother quadrants. The brain’s powerful drivetoward sense making leads us to strive tointerpret our own behavior. Since quadrant Ioften does not have conscious access toactivity in the other quadrants, it is perhapsnot surprising that it tends to over-attributebehavior to itself—i.e., to deliberative deci-sion processes. Even though much of thebrain’s activity is “cognitively inaccessible,”we have the illusion that we are able to makesense of it, and we tend to make sense of itin terms of quadrant I processes.

Research with EEG recordings has shown(Benjamin Libet 1985) that the precisemoment at which we become aware of anintention to perform an action trails the ini-tial wave of brain activity associated with thataction (the EEG “readiness potential”) byabout 300 msec. The overt behavioralresponse itself then follows the sensation ofintention by another 200 msec. Hence, whatis registered in consciousness is a regularpairing of the sensation of intention followedby the overt behavior. Because the neuralactivity antecedent to the intention is inac-cessible to consciousness, we experience“free will” (i.e., we cannot identify anythingthat is causing the feeling of intention).Because the behavior reliably follows theintention, we feel that this “freely willed”intention is causing the action—but in fact,

17 “… the brain contains a specific cognitive system thatbinds intentional actions to their effects to construct acoherent experience of our own agency” (Patrick Haggard,Sam Clark, and Jeri Kalogeras 2002).

both the sensation of intention and the overtaction are caused by prior neural eventswhich are inaccessible to consciousness (seeDaniel Wegner and Thalia Wheatley 1999).17

Quadrant I tends to explain behavior ego-centrically—to attribute it to the types ofdeliberative processes that it is responsiblefor (see Richard Nisbett and Wilson 1977). Adramatic study demonstrating this phenom-enon was conducted with a “split-brain”patient (who had an operation separating theconnection between the two hemispheres ofhis brain). The patient’s right hemispherecould interpret language but not speak, andthe left hemisphere could speak (LeDoux1996). The patient’s right hemisphere wasinstructed to wave his hand (by showing theword “wave” on the left part of a visualscreen, which only the right hemisphereprocessed). The left hemisphere saw theright hand waving but was unaware of theinstructions that had been given to the righthemisphere (because the cross-hemisphereconnections were severed). When thepatient was asked why he waved, the lefthemisphere (acting as spokesperson for theentire body) invariably came up with a plau-sible explanation, like “I saw somebody Iknew and waved at them.”

4. General Implications of Neuroscience for Economics

To add value to economics, neuroscienceneeds to suggest new insights and usefulperspectives on old problems. This sectiondiscusses some broad implications for eco-nomics of the ideas and findings reviewed inthe previous section. First, we show thatneuroscience findings raise questions aboutthe usefulness of some of the most commonconstructs that economists commonly use,such as risk aversion, time preference, and



altruism. Second, we show how the exis-tence of specialized systems challengesstandard assumptions about human infor-mation processing and suggests that intelli-gence and its opposite—boundedrationality—are likely to be highly domain-specific. Third, brain-scans conducted whilepeople win or lose money suggest thatmoney activates similar reward areas as doother “primary reinforcers” like food anddrugs, which implies that money confersdirect utility, rather than simply being val-ued only for what it can buy. Fourth, weshow that research on the motivational andpleasure systems of the brain human chal-lenges the assumed connection betweenmotivation and pleasure. Finally, wedescribe some of the important implicationsof cognitive inaccessibility for economics.

4.1 Economic Constructs

Knowing how the brain solves problems,and what specialized systems it has at its dis-posal to do so, challenges some of our fun-damental assumptions about how peoplediffer from one-another when it comes toeconomic behavior. Economists currentlyclassify individuals on such dimensions as“time preference,” “risk preference,” and“altruism.” These are seen as characteristicsthat are stable within an individual over timeand consistent across activities; someonewho is risk-seeking in one domain is expect-ed to be risk-seeking in other domains aswell. But empirical evidence shows that risk-taking, time discounting, and altruism arevery weakly correlated or uncorrelatedacross situations. This inconsistency resultsin part from the fact that preferences arestate-contingent (and that people may notrecognize the state-contingency, which—ifthey did—would trigger overrides thatimpose more consistency than observed).But it also may point to fundamental prob-lems with the constructs that we use todefine how people differ from each other.

As an illustration, take the concept oftime-preference. In empirical applications,

economic analyses typically assume that thesame degree of time preference is presentfor all intertemporal tradeoffs— saving forthe future, flossing your teeth, dieting, andgetting a tattoo. For example, whether a per-son smokes is sometimes taken as a crudeproxy for low rates of time discounting, instudies of educational investment or savings.Thinking about the modularity of the brainsuggests that while different intertemporaltradeoffs may have some element of plan-ning in common (e.g., activity in prefrontalcortex), different types of intertemporalchoices are likely to invoke qualitatively dif-ferent mixtures of neural systems and henceto produce entirely different patterns ofbehavior. Then measured discount rates willnot be perfectly correlated across domains,and might hardly be correlated at all.

Much as the study of memory has beenrefined by the identification of distinct mem-ory systems, each with its own properties oflearning, forgetting, and retrieval; the study ofintertemporal choice might be enhanced by asimilar decomposition, most likely informedby neuroscience research. For example,unpublished research by Loewenstein andRoberto Weber suggests that, in normal sam-ples, future-oriented behaviors tend to clus-ter in tasks which tap different dimensions ofself-control. For example, flossing your teethis statistically associated with a number ofother minor, repetitive, helpful behaviorssuch as feeding parking meters religiouslyand being on time for appointments. Thesebehaviors seem to involve “conscientious-ness,” an important measure in personalitytheory.18 John Ameriks, Andrew Caplin, andJohn Leahy (in press) measured the propen-sity to plan by asking TIAA–CREF enrolleesto disagree or agree with statements like “Ihave spent a great deal of time developing afinancial plan.” These measures correlatewith actual savings rates.

Dieting and use of addictive drugs, incontrast to punctuality and flossing, mightbe involve very different circuitry andhence reveal fundamentally different dis-



18 Based on the observation that drugs affecting sero-tonin receptors are used to treat compulsion disorders,compulsivity appears to have some connection to the neu-rotransmitter serotonin, In addition, compulsion disordersare thought to be related to hyperactivity in the caudatenucleus, a “reminder” region of the limbic system.

counting behavior. Both of these activitiesinvolve visceral motives that seem to differreliably across persons and may be onlyweakly linked to conscientiousness. Ifyou’re the kind of person who loves to eat,or drink alcohol, then resisting indulging inthose activities requires difficult exertion ofself-control.

4.2 Domain-Specific Expertise

Economics implicitly assumes that peoplehave general cognitive capabilities that canbe applied to any type of problem and,hence, that people will perform equivalentlyon problems that have similar structure. Theexistence of systems that evolved to performspecific functions, in contrast, suggests thatperformance will depend critically onwhether a problem that one confronts canbe, and is in fact, processed by a specializedsystem that is well adapted to that form ofprocessing. When a specialized system existsand is applied to a particular task, processingis rapid and the task feels relatively effort-less. Automatic processes involved in vision,for example, are lightning fast and occurwith no feeling of mental effort, so peopleare not aware of the power and sophistica-tion of the processes that allow it to happen.Even the most powerful computers don’thold a candle to humans when it comes tovisual perception or voice recognition.

When we lack such tailored systems, how-ever, we are likely to seem extraordinarilyflat-footed because we will be forced to“muscle it out” with quadrant 1 processing,much as autistic individuals seem to solvetheory of mind problems by building up astatistical understanding of appropriatesocial behavior. As a general rule, we shouldexpect people to be geniuses when present-ed with problems that can be, and are

processed by specialized systems, but rela-tively obtuse when they are forced to rely oncontrolled processes.

A neat illustration of this is provided by the“Wason four-card problem” in logic. Subjectsare shown four cards, each with a letter onone side and a number on the other. Theexposed card faces read “X,” “Y,” “1,” and “2.”Subjects are asked which cards would need tobe turned over to test the rule: “If there is anX on one side there is a 2 on the other.” Fewsubjects give the right answers, which are Xand 1 (if there’s an X on the opposite side ofthe “1” the rule is broken). However, mostsubjects give the right answer when the logi-cally equivalent problem is put in a cheating-detection frame. For example, if there arefour children from two different towns andtwo school districts and the rule is “If a childlives in Concord he or she must go toConcord High,” most subjects realize that thehome address of the student who does not goto Concord High must be checked to see ifshe is cheating and not going to ConcordHigh when she should (Cosmides 1989).

Of particular interest to economics, manyneuroscientists believe there is a specialized“mentalizing” (or “theory of mind”) module,which controls a person’s inferences aboutwhat other people believe, feel, or mightdo. The hint at the existence of such a ded-icated module came from tests conductedby developmental psychologists in whichtwo children are shown an object in theprocess of being concealed (see Uta Frith2001a). One child then leaves, and theother child observes as the object is movedto a new location. The child who remains inthe room is then asked to predict where thechild who left will look for the object whenshe returns. Normal children are typicallyable to solve this problem around age threeor four. Autistic children as a rule masterthis distinction much later (8–12 years), andwith great difficulty, although some (espe-cially those with “Asperger syndrome”)have normal or superior intelligence. At thesame time, the autistic child will have no



difficulty with general inferences having asimilar logical form (e.g., if a photo is takenof the object’s location, and the object isthen moved, they will correctly infer thatthe photo will register the object in the oldplace, before it was moved).

Autistic adult individuals may compensatein many ways and eventually pass such basictests of mentalizing. However, they have dif-ficulty appreciating more subtle socialmeanings (e.g., irony), and will sometimeswonder at the “uncanny ability” ofnon–autistic persons to “read minds” (Frith2001b). People with Asperger’s syndrome,who are intelligent but have trouble under-standing emotions in others, show loweractivation in the medial prefrontal regions ascompared with normals when presentedwith problems that involve mentalizing, butalso show greater activation in more ventral(lower) region of the prefrontal cortex,which is normally responsible for generalreasoning (Francesca Happe et al. 1996). Anatural interpretation is that Asperger indi-viduals eventually deduce the answerthrough a complex process of reasoninginstead of grasping it directly by a special-ized module, as if they create a neural“workaround” or “long-cut.” In the medicalliterature, one can also find patients withbrain lesions who exhibit difficulties withmentalizing tasks but not with other types ofcognition (Andrea Rowe et al. 2001; JamesBlair and Lisa Cipolotti 2000). This, too, isconsistent with the hypothesis of a separablementalizing module.

The possibility of a mentalizing modulehas gained credibility and substance throughconverging neuroscientific evidence. fMRIstudies have shown that when normal adultsare given pairs of closely matching judgmentproblems, differing only in whether they door do not require mentalizing, the mentaliz-ing problems lead to greater activation in theleft medial prefrontal cortex (Fletcher et al.1995; Rebecca Saxe and Nancy Kanwisher2003). As ultimate proof, one would like toidentify neuronal populations that are

specifically turned on by mentalizing activi-ty. Neuroscience is not there yet, but recentsingle-cell recordings in monkeys have iden-tified an intriguing class of “mirror neurons”in the prefrontal cortex, which fire eitherwhen an experimenter performs a physicalaction (e.g., grasping a peanut) or when themonkey performs (“mirrors”) the sameaction. Having such neurons makes learningby imitation easy and supports mind readingby, for example, internally simulating thefacial expressions of others.

Mentalizing is relevant for economicsbecause many judgments require agents tomake guesses about how other people feel orwhat they will do. The concept of equilibri-um requires that agents correctly anticipatewhat others will do; presumably this arisesbecause of accurate mentalizing, or throughsome kind of specific learning about behav-ior which may not transfer well to newdomains or when variables change.Furthermore, the kind of learning aboutplayers’ “types” from observation inBayesian–Nash equilibrium is modeled assimply Bayesian updating of the likelihoodsof chance events in the face of new informa-tion. Since mentalizing is a special ability,and logical-deductive reasoning can onlypartially compensate for its absence, treatinginference about behavior of other agents and“nature” is a simplification which may bewrong.19

19 Our point here is simple: Circuitry that controlsupdating of frequencies of events in the world—whether itsnows in Chicago in January—may be dissociated from cir-cuitry that controls updating of personal “types” frombehavior. The circuitry which controls attributions of pay-off types to people, from their behavior, requires a conceptof how types link to behavior (i.e., what are types?), as wellas updating of types (i.e., reputations) from behavior. Anautist with theory of mind deficits, for example, mightkeep very accurate counts of relative frequencies (in fact,many autists are obsessed with counting objects in specialcategories) but be unable to tell whether a person behavedbadly in a game because of bad intentions or situationalinfluences (such as a bureaucrat who is pressured to stickto rules). So there is no reason to think that updating aboutevent frequencies is necessarily the same as updatingabout personal types.



The existence of domain-specific expertisesuggests that people will appear to begeniuses at some tasks but will seemremarkably flat-footed when dealing withother tasks that may be only superficially dif-ferent. Domain-specific processing hasimportant implications for economics, andspecifically for the organization of labor.Bundling tasks together into jobs, for exam-ple, requires an understanding of which kindof skills are general (useful for many tasks)and which are neurally separate.

4.3 Utility for Money

As discussed earlier, neuroscience canpoint out commonalities between categoriesthat had been viewed as distinct. An exampleof this with important implications for eco-nomics is the utility for money. The canoni-cal economic model assumes that the utilityfor money is indirect—i.e., that money is amere counter, only valued for the goods andservices it can procure. Thus, standard eco-nomics would view, say, the pleasure fromfood or cocaine and the “pleasure” fromobtaining money as two totally different phe-nomena. Neural evidence suggests, however,that the same dopaminergic reward circuitryof the brain in the midbrain (mesolimbic sys-tem) is activated for a wide variety of differ-ent reinforcers (Montague and Berns 2002),including attractive faces (Itzhak Aharon etal. 2001), funny cartoons (Dean Mobbs et al.2003), cultural objects like sports cars(Susanne Erk et al. 2002), drugs (Schultz2002), and money (e.g., Hans Breiter et al.2001; Brian Knutson and Richard Peterson,in press; Delgado et al. 2000). This suggeststhat money provides direct reinforcement.

Figure 4 is a rough guess about the neuralcircuitry of reward (from Schultz 2002). It isuseful as a pictorial reminder that, while neu-roscience (and our review) often emphasizespecific regions which are central in differentkinds of processing, the focus in thinkingabout economic decisions should be on cir-cuitry or systems of regions and how theyinteract. The diagram also shows how frontal

regions (at the top of Figure 4) both receiveinput from “lower” systems (dopaminergicneurons and the amygdala), and also feedback processed information to the striatum.

The idea that many rewards are processedsimilarly in the brain has important implica-tions for economics, which assumes that themarginal utility of money depends on whatmoney buys. Of course, it is possible thatmoney rewards activate the same circuitry assports cars, cocaine, and jokes because thecircuitry is being used to evaluate money interms of the goods that it buys (whichrequires a cortical process that “simulates”the value of those goods internally to thebrain). But it is also possible that moneybecomes what psychologists call a “primaryreinforcer,” which means that people valuemoney without carefully computing whatthey plan to buy with it. Neuroeconomics isnot nearly advanced enough to separate thesetwo roles for midbrain responses to money,but suggests the possibility that the brainvalue of money is only loosely linked to con-sumption utility, which in turn suggests awide range of potential experiments toexplore implications.

An example of how consumption-drivenand primary reinforcement from money canmatter is asset pricing. Since Robert Lucas(1978), many models of stock prices haveassumed that investors care about the utili-ty of consumption they can finance withstock market returns, rather than withreturns per se. Simple models of this sortmake many counterfactual predictions—most famously that the “equity premium,”or marginal return to stocks over bonds,would be much lower than it has actuallybeen if investors only disliked risk becauseof its impact on consumption, as conven-tional models assume. Shlomo Benartzi andRichard Thaler (1995) and NicholasBarberis, Tano Santos, and Ming Huang(2001) have had more success explainingreturns patterns using a model in whichinvestors care directly about stock returns.This alternative assumption fits a brain that



Figure 4. Hypothesized neural circuitry of reward processing (Schultz 2000).

gets a kick out of earning high returns fortheir own sake.

If gaining money provides direct pleasure,then the experience of parting with it isprobably painful. While there is no directevidence that paying is painful, the assump-tion that paying hurts can explain many mar-ket phenomena which are otherwisepuzzling (Drazen Prelec and Loewenstein1998). An example is the effect of payment-neutral pricing schemes on choices.Companies often go to great lengths to dis-guise payments, or reduce their pain.Consumers appear to oversubscribe to flat-rate payment plans for utilities and tele-phone service and health clubs (KennethTrain 1991; Train, Daniel McFadden, andMoshe Ben-Akiva 1987; Della Vigna andUlrike Malmendier 2003, Anja Lambrechtand Bernd Skiera 2004). A flat-rate planeliminates marginal costs and allows con-sumers to enjoy the service without thinkingabout the marginal cost. Similarly, travel

plans are often sold as packages, making itimpossible to compute the cost of the indi-vidual components (hotel, food, transporta-tion). Often components of the package arepresented as “free” (like Microsoft’s internetbrowser) even though the claim is meaning-less from an economic standpoint, given thatthe package is presented on a take-it-or-leave-it basis. One can interpret the appealof ad-hoc currencies, such as frequent-flyermiles, chips in casinos, or the beads used forincidental expenses at all-inclusive resortslike Club Med, as an attempt to reduce thepain-of-payment. The ad-hoc currency,whether miles or beads, feels like “playmoney,” and spending it does not seem toexact the same psychic cost.

In experiments, we have observed a prefer-ence for prepayment for certain items, evenwhere prepayment is financially irrationalbecause it incurs an opportunity cost of lostinterest payments (Prelec and Loewenstein1998). When questioned, respondents claim



to prefer prepayment for, e.g., a vacation tripbecause then they could relax and enjoy thevacation more knowing it was paid up. It is aninteresting question whether part of themotive for owning rather than renting prod-ucts is precisely to create prepayment, and soenjoy consumption without being reminded ofthe cost. The pain-of-paying may also explainwhy we are willing to pay less for a product ifpaying in cash than by credit card. Althoughthere are financial reasons to prefer paying bycredit-card, the size of credit card relative tocash premium that people reveal, in an exper-iment, is much too high (up to 100 percent) tobe rationalized by liquidity preference andother economic considerations (Prelec andDuncan Simester 2001).

4.4 Wanting and Liking

Economists usually view behavior as asearch for pleasure (or, equivalently, escapefrom pain). The subfield of welfare econom-ics, and the entire ability of economists tomake normative statements, is premised onthe idea that giving people what they wantmakes them better off. But, there is consid-erable evidence from neuroscience andother areas of psychology that the motivationto take an action is not always closely tied tohedonic consequences.

Berridge (1996) argues that decision mak-ing involves the interaction of two separate,though overlapping systems, one responsi-ble for pleasure and pain (the “liking” sys-tem), and the other for motivation (the“wanting” system). This challenges the fun-damental supposition in economics that oneonly strives to obtain what one likes.Berridge finds that certain lesions and phar-macological interventions can selectivelyenhance a rat’s willingness to work for afood, without changing the pleasure of eat-ing the food, as measured, admittedly some-what questionably, by the animal’s facialexpression (Animal facial expressions, likethose of humans, provide at least a clueabout whether something tastes good, bador indifferent. In economic language, the

experiments create a situation where theutility of food and disutility of work remainthe same, but the amount of work-for-reward goes up. This implies that it is possi-ble to be motivated to take actions that bringno pleasure.

Berridge believes that the later stages ofmany drug addictions presents prototypicalexamples of situations of what he terms“wanting” without “liking”; drug addictsoften report an absence of pleasure fromtaking the drugs they are addicted to, cou-pled with an irresistible motivation to do so.Other examples of situations in which thereoften seems to be a disconnect betweenone’s motivation to obtain something andthe pleasure one is likely to derive from itare sex and curiosity (Loewenstein 1994).Thus, for example, you can be powerfullymotivated to seek out information, evenwhen you are quite certain that it will makeyou miserable, and can feel quite unmotivat-ed to engage in activities that, at a purelycognitive level, you are quite sure you wouldfind deeply pleasurable.

Economics proceeds on the assumptionthat satisfying people’s wants is a good thing.This assumption depends on knowing thatpeople will like what they want. If likes andwants diverge, this would pose a fundamen-tal challenge to standard welfare economics.Presumably welfare should be based on“liking.” But if we cannot infer what peoplelike from what they want and choose, thenan alternative method for measuring likingis needed, while avoiding an oppressivepaternalism.

4.5 Cognitive Inaccessibility

The fact that people lack introspectiveaccess to the sources of their own judg-ments of behavior, and tend to overattributeboth to controlled processes, has manyimportant implications for economics.When it comes to discriminatory biases, forexample, because people lack introspectiveaccess to the processes that produce suchbiases, they are unable to correct for them



even when they are motivated to makeimpartial judgments and decisions. Indeed,they are very likely to deny that they arebiased and, hence, likely to not even per-ceive the need for such correction. Suchunconscious discrimination may explainwhy otherwise identical job applicationresumes for candidates with statistically“white” rather than “African–American”names have a 50 percent higher chance ofgenerating callbacks from job applicationletters, as shown recently by MarianneBertrand and Mullainathan (2004). It canalso help explain why physicians are so con-vinced that gifts from pharmaceutical com-panies do not bias their prescriptionpractices even though research (and thecontinuing practice of gift giving by thecompanies) suggests that it does (JasonDana and Loewenstein 2003).

A second class of implications is related tothe phenomena of apparent self-deceptionand self-manipulation, where, for instance,economic agents (investors, consumers,entrepreneurs) are overly optimistic abouttheir chances for success. These phenomenahave been richly catalogued by social psy-chologists, beginning with the research onmotivated cognition and cognitive disso-nance in the 1950s. Neuroscience suggeststhat they are all related to chronic cognitiveinaccessibility of automatic brain processes.Attention, for example, is largely controlledby automatic processes, and attention inturn determines what information weabsorb. If attention is chronically drawn toinformation that is favorable to us, we willemerge with an overoptimistic sense of ourabilities and prospects. It is a case of quad-rants III and IV collaborating, without the“adult supervision” provided by quadrant I.

A third class of implications arises from thecognitive inaccessibility of our own motivesfor action. The fact that we are not con-sciously aware of the moment of decision (asshown by Libet’s research, mentioned in theprevious section) strongly suggests that wemay also not understand the reasons why we

choose one way or another. Paradoxically, thiscan be very beneficial. In many situations, anaction may be diagnostic of a good outcome,without having any significant ability to causethat outcome. For instance, participating insome socially desirable activity (e.g., voting ornot littering) may be quite diagnostic of thedesired collective outcome (your candidatewinning, clean streets) without being able tocause that outcome, because the impact ofyour action is negligible. This gives rise to thenotorious “voter’s paradox” in rational choicetheory, which seems to imply that no oneshould ever vote. Fortunately, the distinctionbetween diagnosticity and causality, which isabsolute in the rational-choice model, is quitefuzzy psychologically. There is experimentalevidence that people will cheat on their ownmedical tests so as to “manufacture” a gooddiagnosis (George Quattrone and AmosTversky 1984), and their own personalityinventories, to yield a personality assessmentdiagnostic of success. This can only be possi-ble if the true motive for action — the desireto get “good news” — is hidden from theagent at the moment of choice; if it were nothidden, then awareness that the action wastaken precisely to get the good news wouldinstantly void the diagnostic value of theaction (see Ronit Bodner and Prelec 2003 foran economic self-signaling model that allowsfor noncausal motives on action). Cognitiveinaccessibility prevents this logical short-cir-cuit and dramatically expands the range ofmotives that can influence behavior. Becausethe hedonic system (quadrant IV) is not con-strained by logical considerations, a personwho takes a small step toward a larger virtu-ous objective, such as joining a gym to get fit,or buying a copy of Stephen Hawking’s ABrief History of Time to feel scientifically lit-erate, may experience a feeling of pleasurefrom the small step “as if” they have in factsecured their objective (which they most like-ly will not). At the same time, cognitive inac-cessibility means that quadrant I doesn’t haveto acknowledge that the feeling of pleasurewas in fact the cause of his action.



5. Specific Economic Applications

We now go into greater detail about rami-fications of neuroscience findings for fourspecific topics in economics: intertemporalchoice, decision making under risk anduncertainty, game theory, and labor-marketdiscrimination.

5.1 Intertemporal Choice and Self-Control

The standard perspective in economicsviews intertemporal choice as a trade-off ofutility at different points in time. Individualdifferences in the way that people make thistradeoff are captured by the notion of a dis-count rate—a rate at which people discountfuture utilities as a function of when theyoccur. The notion of discounting, however,gained currency not because of any support-ive evidence, but based only on its conven-ient similarity to financial net present valuecalculations (Loewenstein 1992). Indeed, inthe article that first proposed DU in detail,Samuelson (1937) explicitly questioned itsdescriptive validity, saying “It is completelyarbitrary to assume that the individualbehaves so as to maximize an integral of theform envisaged in [the DU model].”

In fact, more recent empirical research ontime discounting challenges the idea thatpeople discount all future utilities at a con-stant rate (see Frederick, Loewenstein, andO’Donoghue 2002). The notion of time dis-counting, it seems, neither describes thebehavior of individuals nor helps us to classi-fy individuals in a useful fashion. How can anunderstanding of the brain help us to estab-lish a better understanding of intertemporalchoice behavior? Our two central distinc-tions, between affect and cognition, andbetween automatic and controlled processes,both have important ramifications.

5.1.1 Affect and Cognition in IntertemporalChoice

As discussed above, the affective system isdesigned to ensure that certain survival andreproduction functions are met and it

achieves this function in part by motivatingindividuals to take certain actions. In most ani-mals, emotions and drives motivate behaviorsthat have short-term goals, such as eating,drinking, and copulating. As a result, theaffective systems that we share with widerange of other animals are inherently myopic.Though some animals display far-sightedbehaviors, such as storing food for winter,these are specialized, preprogrammed, behav-iors that are distinctly different from the typeof spontaneous delay of gratification observedin humans. Humans appear to be uniqueamong animals in terms of caring about, mak-ing immediate sacrifices for, and flexiblyresponding to, desired future consequences.

This capacity to take long-term conse-quences of our behavior into account seemsto be the product of our prefrontal cortex,which, tellingly, is the part of the brain thatis uniquely human (see, e.g., Manuck et al.2003). Patients with damage to prefrontalregions—most famously Phineas Gage,whose injury led to a reformulation of ourunderstanding of the function of the pre-frontal cortex—tend to behave myopically,paying little heed to the delayed conse-quences of their behavior. As Hersh Shefrinand Thaler (1988) suggested many years ago,intertemporal choice can be viewed as asplice of two processes—an impulsive, affec-tive, process and a more far-sighted processguided by the prefrontal cortex.

Recent brain imaging research providessupport for such an account. SamuelMcClure et al. (2004) scanned subjects usingfMRI while they made a series of preferencejudgments between monetary reward optionsthat varied by amount and delay to delivery.For some pairs of choices, the earlier rewardwould be received immediately; for others,both rewards were delayed (though one bymore than the other). Consistent with theidea that intertemporal choice is driven bytwo systems, one more cognitive and onemore deliberative, they found that parts ofthe limbic—i.e., affective—system associatedwith the midbrain dopamine system were



preferentially activated by options involvingimmediately available rewards. In contrast,regions of the lateral prefrontal cortex andposterior parietal cortex—typically viewed asmore cognitive regions—were engaged uni-formly by intertemporal choices irrespectiveof delay. Furthermore, the relative activity ofthe two systems actually predicted the choic-es that subjects made: Greater relative activ-ity in affective systems was associated withchoosing earlier rewards more often.

The notion of quasi-hyperbolic time dis-counting, of course, provides a mathematicalrepresentation of precisely such a splicing oftwo processes, and has been shown to use-fully describe behavior in a wide range ofdomains. However, an understanding of theneural underpinnings of these dual process-es allows for more nuanced predictions andformulations. The notion of hyperbolic time-discounting predicts that people will alwaysbehave impulsively when faced with theright combination of incentives (typicallythose involving some immediate cost andbenefit), but this does not seem to be thecase. Understanding that hyperbolic timediscounting stems, in part, from competitionbetween the affective and cognitive systems,leads to the prediction that factors thatstrengthen or weaken one or the other ofthese influences will cause people to behavemore or less impulsively.

5.1.2 Determinants of the Relative Strengthof Affect and Cognition

There are a variety of factors that affect therelative strength of affective and cognitiveinfluences on intertemporal choice whichcan help to explain what could be called“intraindividual” variability in impatience.

First, any factor that increases the demandson the prefrontal cortex—on the controlled,cognitive, system, should decrease the influ-ence of this system and, hence, decrease indi-viduals’ control over their own behavior. Thispossibility was demonstrated by Baba Shivand Alexander Fedorikhin (1999). To manip-ulate “cognitive load,” half of their subjects

(low-load) were given a two-digit number thatthey were instructed to memorize, and theother half (high load) were given a seven-digitnumber to memorize. Subjects were theninstructed to walk to another room in thebuilding. On the way they passed by a table atwhich they were presented with a choicebetween a caloric slice of cake or a bowl offruit salad. More than half (59 percent) chosethe cake in the high load (seven-digit) condi-tion, but only 37 percent chose the cake in thelow load (two-digit number) condition. Thisfinding is consistent with the idea that theeffort required to memorize seven-digit num-bers drew deliberative resources away fromself-control, leading those subjects who hadto remember more to eat more cake.

Second, prior exercise of self-controlseems to diminish the capacity and, hence,propensity to exert self-control in the present.Recall that the prefrontal cortex is the part ofthe brain that is associated with a subjectivefeeling of effort. It is tempting to attributethis to the fact that self-control involves thesame part of the brain—the executive pre-frontal cortex—that is itself associated withfeelings of mental effort. Perhaps this is whyexercising willpower feels so difficult, andwhy exercising self-control in one domain canundermine its exercise in another, as demon-strated by a series of clever experiments con-ducted by Baumeister and colleagues (see,e.g., Baumeister and Kathleen Vohs 2003). Ina typical study, subjects on diets who resistedtemptation (by foregoing the chance to grabsnacks from a nearby basket) later ate moreice cream in an ice-cream taste test and alsoquit earlier when confronted with an intellec-tual problem they couldn’t solve. They actedas if their ability to resist temptation was tem-porarily “used up” by resisting the snacks (or,alternatively, that they had “earned” a rewardof ice cream by skipping the temptingsnacks). Other factors that seem to under-mine this self-control resource are alcohol,stress, and sleep deprivation.

Turning to the other side of the equation,activation of affective states should, by the



same token, tend to accentuate temporalmyopia. Indeed, there is considerable evi-dence of such effects (Janet Metcalfe andMischel 1999). Research has shown, forexample, that addicts display higher discountrates, not only for drugs but also for money,when they are currently craving drugs thanwhen they are not (Giordano et al. 2002).Other research has shown that sexual arousalproduces greater time-discounting of moneyrewards (Ariely and Loewenstein 2003).

The main exceptions to the rule that affec-tive states tend to engender short-sightedbehavior involve interactions between thecognitive and affective systems. In fact, deci-sions to delay gratification often involve amixture of affect and cognition. Theyrequire a cognitive awareness of the delayedbenefits in delaying gratification—e.g., thatdesisting from eating cake today will mean amore pleasing body type in the future. But,as many researchers have observed, cogni-tive awareness alone is insufficient to moti-vate delay of gratification; emotions play acritical role in forward-looking decisionmaking. As Barlow (1988) notes, “The capac-ity to experience anxiety and the capacity toplan, are two sides of the same coin.”

Thomas Cottle and Stephen Klineberg(1974), similarly argue that people careabout the delayed consequences of theirdecisions only to the degree that contem-plating such consequences evokes immedi-ate affect. In support of this view, they citethe effects of frontal lobotomies which harmareas of the brain that underlie the capacityfor images of absent events to generateexperiences of pleasure or discomfort. Theneurosurgeons who performed these opera-tions wrote of their frontal-lobotomypatients that: “the capacity for imagination isstill present, and certainly not sufficientlyreduced to render the patients helpless, andaffective responses are often quite lively,[but there is] a separation of one from theother” (Freeman and Watts 1942).

The work of Damasio and colleagues dis-cussed earlier (Bechara, Damasio, Daniel

Tranel, and Damasio 1997; Damasio 1994)lends further credence to this perspective,as does research on psychopaths, who arecharacterized by both emotional deficitswhen it comes to imagining the future andby insensitivity to the future consequences(as well as consequences to others) of theirbehavior (Hervey Cleckley 1941; Hare1965, 1966; David Lykken 1957). Whetherdeliberately, as when one conjures up animage of a “fat self” exposed on the beach,or without conscious intention, self-controloften involves an interaction of affective andcognitive mechanisms.

How might one model intertemporalchoice differently as a result of the insightsfrom neuroscience? First, the neuroscienceresearch points to ways to “unpack” the con-cept of time preference. Clearly, ability tothink about future consequences is impor-tant, which is probably why time preferenceis correlated with measured intelligence(Mischel and Robert Metzner 1962).Second, because people are likely to makemyopic choices when under the influence ofpowerful drives or emotions (Loewenstein1996), this suggests that a key to under-standing impulsivity in individuals might beto understand what types of situations getthem “hot.” Third, we might be tempted tolook for individual differences in what couldbe called “willpower”—i.e., the availabilityof the scarce internal resource that allowspeople to inhibit viscerally driven behaviors(see Loewenstein and O’Donoghue 2004 fora recent two-system model of intertemporalchoice and other economic behaviors).

A model of intertemporal choice that tookaccount of interactions between affect andcognition can help to explain not only impul-sivity, but also why many people have self-control problems of the opposite type ofthose typically examined in the literature—e.g., tightwads who can’t get themselves tospend enough; workaholics who can’t take abreak, and people who, far from losing con-trol in the bedroom, find themselves frus-tratingly unable to do so. All of these patterns



of behavior can easily be explained by thepossibly uniquely human propensity to expe-rience emotions, such as fear, as a result ofthinking about the future. Indeed, it is likelythat one of the main tools that the prefrontalcortex uses to impose self-control whenaffective forces would otherwise favor short-sighted self-destructive behavior is to create“deliberative affect” via directed imagery andthought (Roger Giner-Sorolla 2001).

Such a framework might also help toexplain why people appear so inconsistentwhen their behavior is viewed through thelens of discounted utility. The ability to thinkabout future consequences may not bestrongly correlated with the degree to whichdifferent experiences produce visceral reac-tions, and these in turn might not be corre-lated with an individual’s level of willpower.Indeed, Loewenstein et al. (2001) foundclose to zero correlations between numerousbehaviors that all had an important intertem-poral component, but much higher correla-tions between behaviors that seemed todraw on the same dimension of intertempo-ral choice—e.g., which required suppressionof specific emotions such as anger.

5.1.3 Automatic Processes in IntertemporalChoice

To the extent that intertemporal choice is,in fact, driven by cognitive considerations,much of this cognition does not take the usu-ally assumed form of a weighing of costs andbenefits discounted according to when theyoccur in time. Rather, consistent with the ear-lier claim that people often make decisions viaa two-part process that takes the form of firstasking “what situation am I in?” then continu-ing “how does one behave in such a situation,”much intertemporal choice is driven by auto-matic processes involving pattern-matching,recognition, and categorization.

A pattern of choice that seems to be drivenby such a process is the preference forsequences of outcomes that improve overtime. In one demonstration of this effect,Loewenstein and Prelec (1993) asked some

subjects to choose whether to eat a fancyFrench Restaurant dinner in one month ortwo, and others to choose between thesequence [fancy French this month, eat athome next month] or the same choice inreverse order. A majority of subjects preferredthe isolated French dinner earlier butdeferred the French dinner when it wasembedded in a sequence with eating at home.Frederick and Loewenstein (2000) conducteda study in which subjects were presented witha series of intertemporal choices that wereframed in ways intended to evoke differentconsiderations. For example, in one versionthey asked respondents to allocate pleasurableoutcomes, such as massages, over time. Theyexpected, and found, that the allocation for-mat evoked a choice heuristic that caused peo-ple to spread consumption relatively evenlyover time, implying a preference for flatsequences. In another version, they askedrespondents to state a maximum buying pricefor the Greek-then-French and French-then-Greek sequences, rather than choose betweenthem, anticipating that the mention of moneywould evoke considerations of the time valueof money and, hence, cause subjects to placehigher value on the option that providedgreater value earlier—i.e., the decliningsequences—which is what they found.20

20 Another phenomenon that may be driven by suchautomatic processes is the “diversification bias” (ItamarSimonson 1989). When people choose several alternativesfrom a set, they choose more variety when they choosethem all simultaneously than when they choose themsequentially. This phenomenon has been demonstratedwith snack food, audio pieces, gambles, and lottery tick-ets. Daniel Read and Loewenstein (1995) tested variousexplanations for overdiversification (e.g., they diversify insimultaneous choice to gather information), but conclud-ed that it results from a rule of thumb that that they applywhenever choices are expressed in a fashion that high-lights diversification (see also Thomas Langer and CraigFox 2004). This viewpoint is supported by a study (Read,Loewenstein, and S. Kalyanaraman 1999) in which sub-jects make successive choices between groups of objectswhich are easily categorized or not. When categorizationwas easy (e.g., virtues and vices), subjects diversifiedmore in simultaneous than in sequential choice. Whenthere were multiple competing categorizations, the overdiversification bias disappeared.



In sum, neuroscience points to some defi-ciencies in the way that economists currentlymodel intertemporal choice and also suggestsdirections for future modeling. A somewhatstylized interterpretation of the results justreviewed would be that some intertemporaldecisions are, in fact, well represented by thediscounted utility model—specifically thoseinvolving detailed deliberation but minimalaffect. However, a wide range of otherintertemporal choices are influenced byaffectively “hot” processes such as drives andemotions, or result from processes that auto-matically evoke a response which depends onthe situation. Models which focus on howthese discrepant processes interact are prom-ising (e.g., Bernheim and Rangel 2004;Loewenstein and O’Donoghue 2004).

5.2 Decision-Making under Risk andUncertainty

Both collaboration and competitionbetween affect and cognition, and betweencontrolled and automatic processes, can alsobe seen in the domain of decision makingunder risk and uncertainty.

5.2.1 Affect versus Cognition

The expected utility model views decisionmaking under uncertainty as a tradeoff ofutility under different states of nature—i.e.,different possible scenarios. But, much asthey do toward delayed outcomes, peoplereact to risks at two different levels. On theone hand, as posited by traditional econom-ic theories and consistent with quadrant I oftable 1, people do attempt to evaluate theobjective level of risk that different hazardscould pose. On the other hand, and consis-tent with quadrant IV, people also react torisks at an emotional level, and these emo-tional reactions can powerfully influencetheir behavior (Loewenstein, Weber,Christopher Hsee, and Ned Welch 2001).

The existence of separate affective andcognitive systems that respond differently torisks is most salient when the two systemsclash. People are often “of two minds” when

it comes to risks; we drive (or wish we weredriving as we sit white-knuckled in our air-plane seat) when we know at a cognitive levelthat it is safer to fly. We fear terrorism, whenred meat poses a much greater risk of mor-tality. And, when it comes to asking someoneout on a date, getting up to speak at the podi-um, or taking an important exam, our delib-erative self uses diverse tactics to get us totake risks, or to perform in the face of risks,that our visceral self would much prefer toavoid. Perhaps the most dramatic illustra-tions of the separation of visceral reactionsand cognitive evaluations, however, comesfrom the phobias that so many people sufferfrom; the very hallmark of a phobia is to beunable to face a risk that one recognizes,objectively, to be harmless. Moreover, fearunleashes preprogrammed sequences ofbehavior that aren’t always beneficial. Thus,when fear becomes too intense it can pro-duce counterproductive responses such asfreezing, panicking, or “dry-mouth” whenspeaking in public. The fact that people payfor therapy to deal with their fears, and takedrugs (including alcohol) to overcome them,can be viewed as further “evidence” that peo-ple, or more accurately, people’s deliberativeselves, are not at peace with their visceralreactions to risks.

5.2.2 Affective Reactions to Uncertainty

A lot is known about the neural processesunderlying affective responses to risks.Much risk averse behavior is driven byimmediate fear responses to risks, and fear,in turn, seems to be largely traceable to theamygdala. The amygdala constantly scansincoming stimuli for indications of potentialthreat and responds to inputs both fromautomatic and controlled processes in thebrain. Patrik Vuilleumier et al. (2001)observed equivalent amygdala activation inresponse to fearful faces that were visuallyattended to or in the peripheral region whichfalls outside of conscious perception (cf. deBeatrice de Gelder, Jean Vroomen, GillesPourtois, and Lawrence Weiskrantz 1999;



Nouchine Hadjikhani and de Gelder 2003;LeDoux 1996; John Morris, C. Buchel, andRaymond Dolan 2001; Whalen et al. 1998).But the amygdala also receives corticalinputs, which can moderate or even overrideits automatic quadrant IV response.

In a paradigmatic experiment that illus-trates cortical overriding of amygdala activa-tion (LeDoux 1996), an animal such as a ratis “fear-conditioned”—by repeatedly admin-istering a signal such as a tone followed byadministration of a painful electric shock.Once the tone becomes associated in theanimal’s mind with the shock, the animalresponds to the tone by jumping or showingother over signs of fear. In the next phase ofthe experiment, the tone is played repeated-ly without administering the shock, until thefear response becomes gradually “extin-guished.” At this point, one might think thatthe animal has “unlearned” the connectionbetween the tone and the shock, but thereality is more complicated and interesting.If the neural connections between the cortexand the amygdala are then severed, the orig-inal fear response to the tone reappears,which shows that fear conditioning is noterased in “extinction” but is suppressed bythe cortex and remains latent in the amyg-dala. This suggests that fear learning may bepermanent, which could be an evolutionarilyuseful adaptation because it permits a rapidrelearning if the original cause of the fearreappears. Indeed, other studies (reportedin the same book) show that fear-condition-ing can be reinstated by the administrationof a single shock.

Decision making under risk and uncer-tainty, like intertemporal choice, nicely illus-trates both collaboration and competitionbetween systems. When it comes to collabo-ration, risk taking (or avoiding) behaviorinvolves an exquisite interplay of cognitiveand affective processes. In a well-knownstudy that illustrates such collaboration(Bechara et al. 1997), patients suffering pre-frontal damage (which, as discussed above,produces a disconnect between cognitive

and affective systems) and normal subjectschose a sequence of cards from four deckswhose payoffs the subjects only learned fromexperience (a “multiarmed bandit” prob-lem). Two decks had more cards withextreme wins and losses (and negativeexpected value); two decks had less extremeoutcomes but positive expected value. Bothgroups exhibited similar skin conductance(sweating—an indication of fear) after large-loss cards were encountered, but, comparedto normals, prefrontal subjects rapidlyreturned to the high-paying risky decks aftersuffering a loss and, as a result, went “bank-rupt” more often. Although the immediateemotional reaction of the prefrontal patientsto losses was the same as the reaction of nor-mals (measured by skin conductance), thedamaged patients apparently do not storethe pain of remembered losses as well as nor-mals, so their skin conductance rose muchless than normals when they resampled thehigh risk decks. Subsequent research found asimilar difference between normal subjectswho were either high or low in terms of emo-tional reactivity to negative events. Thosewho were more reactive were more prone tosample from the lower-paying, safer decks ofcards (Peters and Slovic 2000).

Damasio et al.’s research shows thatinsufficient fear can produce nonmaximiz-ing behavior when risky options have nega-tive value. But, it is well established thatfear can also discourage people from takingadvantageous gambles (see, e.g., UriGneezy and Jan Potters 1997). Indeed, Shivet al. (2005) found that frontal patientsactually make more money on a task inwhich negative emotions cause normal sub-jects to be extremely risk averse: a series oftake-it-or-leave-it choices to play a gamblewith a 50 percent chance of losing $1.00 orgaining $1.50. Normal subjects and frontalsubjects were about equally likely to playthe gamble on the first round, but normalsstopped playing when they experiencedlosses, while frontal patients kept playing.Clearly, having frontal damage undermines



the overall quality of decision making; butthere are situations in which frontal damagecan result in superior decisions.

At a more macro level, emotional reac-tions to risk can help to explain risk-seekingas well as risk-aversion (Caplin and Leahy2001). Thus, when gambling is pleasurable, amodel that incorporates affect naturally pre-dicts that people will be risk-seeking andthat self-control will be required to rein inrisk-taking. Indeed, about 1 percent of thepeople who gamble are diagnosed as “patho-logical”—they report losing control, “chasinglosses,” and harming their personal and workrelationships by gambling (NationalAcademy of Sciences 1999). The standardeconomic explanations for gambling—con-vex utility for money or a special taste for theact of gambling—don’t help explain whysome gamblers binge and don’t usefullyinform policies to regulate availability ofgambling. Neuroscience may help.Pathological gamblers tend to be over-whelmingly male and tend to also drink,smoke, and use drugs much more frequent-ly than average. Genetic evidence shows thata certain gene allele (D2A1), which causesgamblers to seek larger and larger thrills toget modest jolts of pleasure, is more likely tobe present in pathological gamblers than innormal people (David Comings 1998). Onestudy shows tentatively that treatment withnaltrexone, a drug that blocks the operationof opiate receptors in the brain, reduces theurge to gamble (e.g., Paula Moreyra et al.2000). The same drug has been used to suc-cessfully treat “compulsive shopping” (SusanMcElroy et al. 1991).

Neural evidence also substantiates the dis-tinction between risk (known probability)and “Knightian” uncertainty, or ambiguity.Subjects facing ambiguous gambles—know-ing they lack information they would like tohave about the odds—often report a feelingof discomfort or mild fear. Brain imagingshows that different degrees of risk anduncertainty activate different areas of thebrain (McCabe et al. 2001; Aldo Rustichini

et al. 2002) which corroborates the subjects’self-reports. Using fMRI, Ming Hsu et al.(2005) found frontal insula and amygdalaactivation when subjects faced ambiguouschoices, compared to risky ones. They alsofound that patients with orbitofrontal corti-cal (OFC) lesions are ambiguity-neutral,compared to brain-damaged controls. Sincethe OFC receives input from the limbic sys-tem (including the insula and amygdala), thefMRI and lesion evidence together implythat in normal subjects, ambiguous gamblesoften create discomfort or fear which istransmitted to the OFC. Ironically, patientswith OFC brain damage therefore behavemore “rationally” (treating ambiguous andrisky gambles similarly) than normals, areminder that logical principles of rationalityand biological adaptations can be different(cf. Shiv et al. 2005).

5.2.3 Automatic Versus ControlledProcesses

The divergence between different sys-tems’ evaluations of risk can also be seenwhen it comes to judgments of probability.Numerous studies by psychologists haveobserved systematic divergences betweenexplicit judgments of probability in differentsettings (presumably the product of con-trolled processing) and implicit judgmentsor judgments derived from choice (whichare more closely associated with automaticprocessing and/or emotion). For example,Kirkpatrick and Epstein (1992) found thatpeople prefer to draw a bean from a bowlcontaining ten winning beans and ninetylosing beans than from a bowl containingone winning bean and nine losing beans (seealso Veronika Denes-Raj, Epstein, andJonathon Cole 1995; Paul Windschitl andGary Wells 1998). Subjects say that theyknow the probabilities of winning are thesame, but they still have an automatic quad-rant III preference for the bowl with morewinning beans.

An important feature of good probabilityjudgment is logical coherence: probabilities



of mutually exclusive and exhaustive eventsshould add to one, and conditional probabil-ities should be linked to joint and marginalprobability according to Bayes rule(P(A|B)=P(A and B)/P(B)). Logical coher-ence is violated in at least two neurallyinteresting ways. One is “conjunction falla-cy”—the tendency to judge events with twocomponents A and B as more likely than Aor B alone. While most subjects (even statis-tically sophisticated ones) make conjunctionerrors on some problems, when those errorsare pointed out, quadrant I wakes up, andthe subjects sheepishly recognize the errorand correct it (Kahneman and Frederick2002). For example, the famous “Linda”problem describes an earnest young politi-cally minded student. In one condition, sub-jects are asked to rank statements aboutLinda—Is she a bank teller? A feminist bankteller? A large majority of subjects, evenhighly educated ones, say Linda is morelikely to be a feminist bank teller than she isto be a bank teller, a violation of the con-junction principle. But when subjects areasked, “Out of one hundred people likeLinda, how many are bank tellers? Feministbank tellers?” conjunction errors disappear(Tversky and Kahneman 1983).

Another violation is that subjects oftenreport probabilities which are logically inco-herent. fMRI evidence suggests an explana-tion for why probability judgments areincoherent, but can be corrected uponreflection: when guessing probabilities, theleft hemisphere of the brain is more active;but when answering logic questions, theright hemisphere is more active (LawrenceParsons and Daniel Osherson 2001). Sinceenforcing logical coherence requires theright hemisphere to “check the work” of theleft hemisphere, there is room for slippage.

Once again, it can be seen that neuro-science, and specifically, a consideration ofaffective and automatic processes that havebeen largely neglected by economists, couldpotentially inform an important line ofresearch and theory. We are skeptical of

21 Of course, equilibration might occur through aprocess other than introspection—e.g., adaptive learning,imitation, communication, or evolution. But these processesmust have a neural basis too.

whether any theory that fails to incorporatethe affective dimensions of risk will be capa-ble of shedding much light on such importantphenomena as stock market booms and busts,the ubiquity of gambling (e.g., slot machinesrevenues dwarf revenues from movies), andthe vicissitudes of public responses to threatsas diverse as terrorism and global warming.

5.3 Game Theory

Neuroscientific data are well-suited toexploring the central assumptions on whichgame theory predictions rest. These assump-tions are that players: (1) have accuratebeliefs about what others will do (i.e., play-ers are in equilibrium); (2) have no emotionsor concern about how much others earn (auseful auxiliary assumption); (3) plan ahead;and (4) learn from experience.

5.3.1 Theory of Mind and Autism

In strategic interactions (games), knowinghow another person thinks, and how anoth-er person thinks you think, etc., is critical topredicting the other person’s behavior (andfor inferring the other player’s intentions,which underlie emotional judgments of fair-ness and obliged reciprocity in more mod-ern theories (e.g., Rabin 1993). From aneural view, iterated strategic thinking con-sumes scarce working memory and alsorequires a player to put herself in anotherplayer’s “mind.” There may be no generichuman capacity to iterate this kind of think-ing beyond a couple of steps.21 Studies thatexamine either subject’s choices, or thatmonitor what type of information subjectslook up or pay attention to in experimentalgames, suggest only one–two steps of strate-gic thinking are typical in most populations(e.g., Eric Johnson et al. 2002; MiguelCosta-Gomes et al. 2001; Camerer, TeckHo, and Kuan Chong 2004), though up to



22 Grether et al. (2004) also find BA 10 activity, alongwith interesting activity in anterior cingulate and the basalforebrain, as subjects bid in second-price auctions.

three–four steps are observed in analyticallyskilled and specially trained populations.

As discussed in section 4.2, many neuro-scientists believe there is a specialized“mind-reading” (or “theory of mind”) area,perhaps in prefrontal area Brodmann 10,which generates reasoning about what oth-ers believe and might do (e.g., Simon Baron-Cohen 2000).22 Autism is thought to be adeficit in this area (and related circuitry).People with autism often have trouble figur-ing out what other people think and believe,and are consequently puzzled by behaviorthat most people would consider normal.

One tool used in behavioral game theory isthe “ultimatum game.” In this game, a “pro-poser” offers a division of a sum of money,generically $10.00, to another “responder”who can accept or reject it, ending the game.If the responder has no emotional reactionto the fact that the proposer is earning morethan she is (“envy” or “disgust”), then theresponder should accept the smallest offer.If the proposer also has no emotional reac-tion to earning more (“guilt”) and anticipatescorrectly what the responder will do, thenthe proposer should offer the lowestamount. This pattern is rarely observed:Instead, in most populations the proposeroffers 40–50 percent and about half theresponders reject offers less than 20 percent.

When players do follow the dictates ofgame theory, the result can be a low payoffand confusion. Consider this quote from anupset subject, an Israeli college student,whose low offer in a $10.00 ultimatum gamewas rejected (from Shmuel Zamir 2000):

I did not earn any money because all the otherplayers are stupid! How can you reject a positiveamount of money and prefer to get zero? Theyjust did not understand the game! You shouldhave stopped the experiment and explained it tothem . . .

Ironically, while the subject’s reasoningmatches exactly how conventional gametheory approaches the game, it also soundsautistic, because this subject is surprisedand perplexed by how normal peoplebehave.

This anecdote is complemented byElizabeth Hill and David Sally’s (2003)extensive comparison of normal and autisticchildren and adults playing ultimatumgames. Nearly half of the autistic childrenoffer zero or one unit (out of ten) in the ulti-matum game, and relatively few offer half.Many autistic adults also offered nothing,but a large number of autistic adults offeredhalf—as if they had developed a reasoningor experiential workaround which tellsthem what other people think is fair ingames that involve sharing, even thoughthey cannot guess what others will do usingnormal circuitry.

McCabe et al. (2001) used fMRI to meas-ure brain activity when subjects playedgames involving trust, cooperation, and pun-ishment. They found that players who coop-erated more often with others showedincreased activation in Brodmann area 10(thought to be one part of the mind-readingcircuitry) and in the thalamus (part of theemotional “limbic” system). Players whocooperated less often showed no systematicactivation.

Bhatt and Camerer (in press) used fMRIto compare brain activity when subjectsmake choices and express beliefs in matrixgames. They found that when players werein equilibrium (choices were best-responsesand beliefs were accurate), the same circuit-ry was being used when making a choiceand expressing a belief. This means equilib-rium is a “state of mind,” which can be iden-tified by a tight overlap in activity in the twotasks, as well as a mathematical restrictionon best response and belief accuracy. Theyalso found evidence suggesting that, whensubjects guessed what beliefs other subjectshad about their own behavior, they tendedto anchor on their own choices. This means



23 The ACC also contains a large concentration of“spindle cells”—large neurons shaped like spindles,which are almost unique to human brains (Allman et al.2002). Loosely speaking, these cells are probably impor-tant for many of the activities which distinguish humansfrom our primate cousins, particularly language, andcomplex decision making.

beliefs about beliefs are not just iterationsof some belief processing mechanism;instead, self-referential beliefs use compo-nents of circuitry for forming beliefs and formaking choices.

5.3.2 Emotions and Visceral Effects

One of the most striking neuroscientificfindings about game theory comes fromSanfey et al.’s (2003) fMRI study of ultima-tum bargaining. By comparing the brains ofsubjects responding to unfair ($1.00–$2.00out of $10.00) and fair ($4.00–5.00) offers,they found that very unfair offers differen-tially activated three regions: Dorsolateralprefrontal cortex (DLPFC), anterior cingu-late (ACC), and insula cortex (see Figure 4).DLPFC is an area involved in planning. Theinsula cortex is known to be activated duringthe experience of negative emotions likepain and disgust. ACC is an “executive func-tion” area which often receives inputs frommany areas and resolves conflicts amongthem.23 Therefore, it appears that, after anunfair offer, the brain (ACC) struggles toresolve the conflict between wanting toaccept the money because of its plannedreward value (DLPFC) and disliking the“disgust” of being treated unfairly (insula).

In fact, whether players reject unfairoffers or not can be predicted rather reliably(a correlation of 0.45) by the level of theirinsula activity. It is irresistible to speculatethat the insula is a neural locus of the distastefor inequality or unfair treatment posited bymodels of social utility, which have been suc-cessfully used to explain many varying pat-terns in experiments — robust ultimatumrejections, public goods contributions, andtrust and gift-exchange (e.g., Bazerman,Loewenstein, and Leigh Thompson 1989;

24 It also suggests an intriguing follow-up experimentthat no previous theory would have predicted—patientswith damage to their insula regions should feel no disgustand accept low offers, unless those patients have devel-oped a “workaround” or alternative method to “feel”unfairness in nonvisceral terms.

Ernst Fehr and Simon Gachter 2000;Camerer 2003). The fact that unfair offersactivate insula means that a verbal statementlike “I am so disgusted about being treatedthat way” is literal, not metaphorical—theyreally do feel disgusted.24

Zak et al. (2003) explored the role of hor-mones in trust games. In a canonical trustgame, one player can invest up to $10.00,which is tripled. A second “trustee” playercan keep or repay as much of the tripledinvestment as they want. Zak et al. measuredeight hormones at different points in thetrust game. The hormone with the largesteffect was oxytocin—a hormone that risesduring social bonding (such as breast-feed-ing and casual touching). They found thatoxytocin rose in the trustee if the first player“trusts” her by investing a lot. (They alsofound that ovulating women were particular-ly untrustworthy—they did not repay asmuch of the investment.)

Roxanna Gonzalez and Loewenstein(2004) examined the impact of circadianrhythms in a repeated trust (centipede)game. They sorted people into “morning”and “night” people (which can be donerather reliably) and had them play a cen-tipede game when they were on- or off-peak(e.g., the morning people were off-peakwhen playing in the evening). Based on priorresearch showing that sleep-cycle affectsemotional regulation—i.e., people’s ability tosuppress or avoid acting on unwanted feel-ings—they predicted and found much lowerlevels of cooperative behavior when peopleplayed at off-peak times.

Tania Singer et al. (2004) report an impor-tant link between reward and behavior ingames. They played repeated prisoners’dilemma games in which one player, in thefMRI scanner, faced a series of opponents.



Figure 5. Coronal slices showing regions which are differentially active after an unfair offer ($1–2 out of $10), relative to activity after a fair off ($4–5). Regions are anterior cingulate (ACC),

right and left insula, and dorsolateral prefrontal cortex (DLPFC). See Sanfey et al. (2003).

The scanned subject was told that someopponents cooperated intentionally (i.e.,they could choose freely) while others coop-erated, but unintentionally. Afterwards, sub-jects were shown different faces of thosethey had played against. The faces of theintentional cooperators activated insula,amygdala, and ventral striatal areas (amongothers). Since the striatum is an all-purposereward area, activation in that region meansthat simply seeing the face of a person whointentionally cooperated with you is reward-ing. In game theory terms, a person’s “repu-tation” in a repeated game is a perception byother players of their “type” or likely behav-ior based on past play. Singer et al.’s resultsmean that a good reputation may be neural-ly encoded in a way similar to beautiful orother rewarding stimuli.

The facts that insula activity, oxytocin lev-els, and sleep cycles affect behavior in games,and that cooperators’ faces “feel beautiful,”does not “disprove” game theory, per se,because preferences for different outcomes,

and reasoning ability, might legitimately varywith these biological factors. The theory iseasily patched by inserting variables, like “anenvy/disgust coefficient,” which depends onsome biological state. But game theory alsoassumes that players will recognize the state-dependence in others and adjust their guess-es about how other people will play. We haveno idea if they can, and it is likely that peopleare generally limited at simulating emotionalstates of others (see Leaf van Boven,Loewenstein, and David Dunning 2003).

Cognitive inaccessibility also implies thatpeople may not fully understand the influ-ence of exogenous changes in visceralstates on their own behavior. For example,if being trusted produces oxytocin, thenwhen oxytocin surges for exogenous rea-sons—from a relaxing massage, or whensynthetic oxytocin is administered—thebrain might misread this surge in oxytocinas a sign of being trusted and react accord-ingly (e.g., by acting in a more reciprocaltrustworthy way).



5.3.3 Backward Induction

A central principle in game theory is“backward induction” in extensive-form(“tree”) games that are played over time.Backward induction means figuring out whatto do today by and reasoning how others willbehave at all possible future points andworking backward. Behavioral evidence, anddirect evidence from measuring where play-ers look on a computer screen, shows thatpeople have trouble doing more than a cou-ple of steps of backward induction (e.g.,Johnson et al. 2002). However, Johnson et al.also found that, when players were brieflyinstructed about how to do backward induc-tion, they could learn to do it rapidly andwith little effort (total response times weresimilar to in pre-instruction trials). This is areminder, in the game theory context, of theimportant distinction between controlledand automatic behavior stressed in section 3above. Initially, subjects look automaticallyat early periods and hardly notice futureperiods they realize (correctly, in a statisticalsense) are unlikely to occur. However, withinstruction and practice, backward inductionquickly becames automated, yielding fastresponses and few errors.

Economists are naturally inclined tomodel cognition in terms of costs and bene-fits. If forced into this framework, backwardinduction would probably be characterizedas cognitively costly. But the fact that it iseasily learned and automated suggests thecosts of backward induction has a specialstructure— at first, it is unnatural (not spon-taneously intuited by subjects), like a pieceof software not yet installed; but onceinstalled, it is cheap to run.

5.3.4 Learning

The idea that a game-theoretic equilibri-um resulted from learning, imitation, or evo-lution, rather than simple introspection, hasled to a large literature on what results in thelong-run from different types of learningmodels (e.g., Drew Fudenberg and David

Levine 1998; George Mailath 1998). Sincemany types of learning rules have been pro-posed, fitting rules to data from experimenthas proved useful in showing when intuitive-ly appealing rules might fit badly, and sug-gesting improvements to those rules (e.g.,Camerer 2003). Camerer and Ho (1999)showed that both simple reinforcement ofchosen strategies, and learning by updatingbeliefs about other players, are really twopolar opposite types of generalized rein-forcement learning in which strategies havenumerical propensities or attractions whichare adjusted over time by experience.

In neural terms, the Camerer–Ho theorycan be interpreted as a splice of two process-es—a rapid, emotional process in which achosen strategy is quickly reinforced by thegain or loss that resulted and a slower delib-erative process that requires players to cre-ate counterfactuals about how much theywould have earned from other strategies thatwere not chosen. Conventional reinforce-ment learning neglects the second process.“Fictitious play” belief learning assumes thesecond process completely overrides thefirst. A parameter in the theory which repre-sents the relative strength of the secondprocess, compared to the first, is usuallyestimated to be between zero and one. Thisimplies that reinforcement is stronger, butboth processes are at work.

Michael Platt and Paul Glimcher (1999)found corroborating evidence for reinforce-ment learning of the first, rapid-process sortin single-neuron recording in monkey pari-etal cortex. They measured neuron firingrates in advance of choices in a gamebetween a monkey and a computerizedopponent. They found that firing rates areclosely related to the average reinforcementreceived for that choice in the last ten trials.Dominic Barraclough, Michelle Conroy, andDaeyeol Lee (2004) found similar evidenceof “learning neurons” in dorsolateral pre-frontal cortex of rhesus monkeys; theirparameter estimates of learning models alsosupports the two-process Camerer–Ho



Figure 6. ( from Glimcher et al., in press). The left graph (“before EV cue”) shows the LIP neuron firingrate (y axis) plotted against the relative expected value of one of two possible reward-producing locations.The right graph (“after cue”) shows the firing rate after the monkey learns which movement is rewarding.

Before EV Cue

Relative Expected Value

After Cue

Act

ion

Pote

ntia

lspe

r se

cond

25 They estimate two reinforcements: When the mon-keys choose and win (reinforcement by �1), and when theychoose and lose (�2). In their two-strategy games, themodel is equivalent to one in which monkeys are not rein-forced for losing, but the unchosen strategy is reinforcedby �2. The fact that �2 is usually less than �1 in magnitude(see also Lee et al., in press) is equivalent to � � 1 in theCamerer–Ho theory.

theory.25 Gathering neural evidence fromhumans (to see whether a second, delibera-tive process is being used as the parametricestimates suggest) is feasible and couldprove insightful.

Figure 6 illustrates the behavior ofGlimcher’s parietal neurons in a task withtwo targets of different expected value. Theleft graph shows the actual firing rates (onthe y axis) plotted against the expectedreward value of the target before the “best”target is revealed. The correspondence isremarkably close—these neurons are encod-ing expected value. The righthand graphshows that, after the winning target isrevealed (so that the expected value is nolonger relevant), the firing rates increase toclose to their maximum capacity.

In sum, neuroscience provides some newways to think about central elements of game

theory. Strategic thinking is likely to requirespecialized circuitry and iterated strategicreasoning is likely to be bounded (as indicat-ed by ample behavioral data). Early studiessuggest hormones and biological factors(such as circadian rhythyms) play a role insocial preferences like trust. Thinking aboutthe brain—and some experimental facts—also suggest that logical principles like back-ward induction may be neurally unnaturaland that learning processes are likely to be asplice of cognitive and affective processes. Atthe same time, single-neuron monkey meas-urement suggests that expected-value com-putation and reinforcement learning maycontrol behavior in highly adapted tasks. Theemerging picture is one in which the simplestelements of game theory, like keeping trackof what has worked in a mixed-equilibriumgame with limited scope for outguessing one’sopponent, may be alive and well in the brain,while higher-order cognition and affectiveinfluence on social preferences depart fromthe standard ideas we teach our students.

5.4 Labor-Market Discrimination

Our last specific application is labor-mar-ket discrimination. Economic models assume



that labor-market discrimination againstminorities is either a taste (a distaste forworking with minorities, or a distaste passedon from customers), or a belief that minorityworkers are less productive (for example, abelief that minority status is a proxy for unob-servable differences in skill, also known as“statistical discrimination”).

Neuroscience suggests a different answer.Automaticity contributes to discriminationbecause neural networks rapidly spread acti-vation through associated concepts andstereotypes. Affect contributes to discrimina-tion because automatic affective reactionshave such a powerful effect on cognitive judg-ments. Discrimination in this view involvesrapid, automatic, associations between socialcategories, stereotypes, and affect.

Such an account receives support fromremarkable experiments that demonstratedsubtle “implicit associations” between demo-graphic categories and good or bad adjectives(try it out on yourself at http://buster.cs.yale.edu/implicit/). Subjects takingthis computer-administered implicit associa-tion test (IAT) are shown a mixed up series ofstereotypically black or white names (Tyroneor Chip) and positive or negative adjectives(mother or devil). They are asked to tap onekey when they see one type of name or adjec-tive, and a different key if they see the othertype of name or adjective, at which point thecomputer moves on to the next name or adjec-tive. The dependent variable is how long ittakes the subject to work their way throughthe complete list of names and adjectives.White subjects work their way through the listmuch more quickly when one key is linked tothe pair (black or negative) and the other to(white or positive) than they are when one keyis linked to [black or positive] and the other to[white or negative]. What’s going on? Thebrain encodes associations in neural networks,which spread activation to related concepts.For white students, black names are instantlyassociated with negative concepts, whetherthey realize it or not, because the associationis automatic (rapid and unconscious).

As the name suggest, the implicit associa-tion test taps into “implicit,” as opposed to“explicit,” attitudes. One can think ofimplicit attitudes, roughly, as those associat-ed with automatic processing, whereasexplicit attitudes are associated with con-scious, controlled, processing. New meth-ods developed by psychologists, such as theIAT, have begun to reveal that implicit andexplicit attitudes can sometimes divergefrom one-another, with implicit attitudes, insome cases, exerting a more reliable influ-ence on behavior. Thus, in one recent study,Allen McConnell and Jill Leibold (2001)administered the IAT to subjects, had themcomplete measures of explicit attitudestoward blacks and whites, and also had theminteract with two experimenters, one blackand the other white. Coders blind to eitherthe implicit or explicit measures then codedthe subjects’ interactions with the experi-menters, including such objective measuresas how closely the subject moved their chairto the experimenter, and the experimenteralso rated their perception of the prejudiceof the subject. Although the experimentersratings of the subjects’ degree of prejudicecorrelated with both explicit (r = 0.33, p �0.05) and implicit (r = 0.39, p � 0.05) meas-ures, the other behavioral measures of biasthat correlated with either the IAT or self-reports of prejudice (these were: coders’overall ratings, speaking time, smiling,speech errors, speech hesitation, and extentof extemporaneous social comments) allcorrelated with the IAT but not with theexplicit measures of prejudice.

Implicit attitudes have also been linked toneural processing. In one study (ElizabethPhelps et al. 2000), researchers adminis-tered the IAT to Caucasian subjects and alsoasked them explicit questions about theirattitudes toward African–Americans. Thenthey scanned the subjects’ brains with fMRIwhile exposing them to photographs of unfa-miliar black and white males, and focusingspecifically on a subcortical structure in thebrain called the amygdala, which numerous



studies have linked to processing of fear.They found that the relative strength ofamygdala activation to black as comparedwith white faces was correlated with the IATmeasure of implicit attitudes, but not withdirect, conscious, expression of race atti-tudes. (A study by Hart et al. (2000) suggeststhe effect is roughly symmetric for blacksubjects seeing white faces.) Furthermore,the same pattern was not observed when theblack and white faces were well known, pos-itively regarded, celebrities (such as MichaelJordan and Denzel Washington). Peoplehold implicit attitudes involving not onlyrace, but also a wide range of other individ-ual characteristics, such as height, weight,attractiveness, religion, and national origin,and, like race, many of these attitudesundoubtedly have real economic conse-quences. Thus, although most people rejectheight and attractiveness as indicators ofmarginal productivity, taller and more attrac-tive people are more likely to earn higherwages (e.g., Nicola Persico, AndrewPostlewaite, and Dan Silverman 2002) andother rewards (e.g., the U.S. Presidency).

Does the implicit association view implythat labor-market discrimination is due to ataste, a statistical shortcut, or somethingelse? One piece of evidence suggests the sta-tistical interpretation is on the right track:the amygdala that is active when people seeother-race faces seems to be sensitive tofamiliarity of faces, not race per se (Duboiset al. 1999). This is consistent with an inter-pretation of statistical discrimination inwhich employers do not fear minority work-ers, they are simply less sure of their abilities.But other evidence suggests discriminationis “something else.” Amygdala activity can bedampened if subjects are shown pictures ofblack and white faces and asked to judgehow much the pictured people like vegeta-bles (Wheeler and Fiske 2005). This suggeststhat automatic reactions to race can beerased (or substituted for) depending on thequestion being asked when the faces are per-ceived. These reactions respond to variables

other than prices and choice sets so it is astretch to think of them of as conventionaltastes or beliefs.

6. Conclusions

Economics parted company from psychol-ogy in the early twentieth century.Economists became skeptical that basic psy-chological forces could be measured withoutinferring them from behavior (the same posi-tion “behaviorist” psychologists of the 1920sreached), which led to adoption of the usefultautology between unobserved utilities andobserved (revealed) preferences. Butremarkable advances in neuroscience nowmake direct measurement of thoughts andfeelings possible for the first time, openingthe “black box” which is the building block ofany economic interaction and system—thehuman mind.

Most economists are curious about neuro-science, but instinctively skeptical that it cantell us how to do better economics. The tra-dition of ignoring psychological regularity inmaking assumptions in economic theory isso deeply ingrained—and has proved rela-tively successful—that knowing more aboutthe brain seems unnecessary. Economic the-ory will chug along successfully for the nextfew years paying no attention at all to neuro-science (just as it paid little attention to psy-chology until recently). But it is hard tobelieve that some neuroscientific regularitieswill not help explain some extant anomalies,particularly those that have been debatedfor decades.

Indeed, in many areas of economics thereare basic constructs or variables at the heartof current debates which can be usefullythought of as neural processes, and studiedusing fMRI and other tools. For example,finance is a field awash in literally millions ofobservations of daily price movements.Despite having widespread access to terrificdata, after decades of careful research thereis no agreed-upon theory of why stock pricesfluctuate, why people trade, and why there



are so many actively managed mutual fundsdespite poor fund performance. Perhapsknowing more about basic neural mecha-nisms that underlie conformity, attentionpaid to large price changes, wishful thinking,sense-making of random series, and percep-tions of expertise can help explain these puzzles (e.g., Lo and Repin 2002).Furthermore, some scholars have arguedthat large fluctuations in stock prices aredue to reasonable time-variation of risk pre-mia. But there is no theoretical basis infinance for why attitudes toward risk wouldvary over time. Maybe neuroscience cansupply one.

In labor markets, a major puzzle is whywages are so downward sticky. Firms say theyare afraid to cut wages because they want tomaintain worker morale (e.g., TrumanBewley 2002); and experiments show thatwhen worker productivity is valuable, payinga high wage induces effort, even when work-ers are free to shirk (Fehr and Gachter2000). Presumably morale is some combina-tion of workers’ emotional feelings towardtheir employer and may be very sensitive torecent experience, to what other workersthink, to whether wage cuts are “procedural-ly just,” and so forth. There is no reasonthese processes could not be described asneural processes and studied that way.

There are many anomalies in intertempo-ral choice. In the United States, credit carddebt is substantial ($5,000.00 per household)and a million personal bankruptcies havebeen declared in each of several years(Laibson, Andrea Repetto, and JeremyTobacman 1998). Healthier food is cheaperand more widely available than ever before,but spending on dieting and obesity are bothon the rise. Surely understanding how brainmechanisms process reward, and curb orproduce compulsion, and their evolutionaryorigins (e.g., Trent Smith 2003) might helpexplain these facts and shape sensible policyand regulation.

Prevailing models of advertising assumethat ads convey information or signal a

product’s quality or, for “network” or “sta-tus” goods, a product’s likely popularity.Many of these models seem like strainedattempts to explain effects of advertisingwithout incorporating the obvious intuitionthat advertising taps neural circuitry ofreward and desire.

Finally, economic models do not provide asatisfying theory of how individuals differ. Aslaymen, we characterize other people asimpulsive or deliberate, stable or neurotic,decisive or indecisive, mature or immature,foolish or wise, depressed or optimistic, scat-terbrained or compulsively organized. Theconsumers who spend countless dollars onself-help, “organize your life” manuals, andwho sustain the huge, and infinitely variedpsychological counseling industry, are typi-cally unhappy where they stand on some ofthese dimensions and are looking for ways tochange. Comparative economic develop-ment, entrepreneurial initiative and innova-tion, business cycle sensitivity, and otherimportant macroeconomic behaviors areprobably sensitive to the distribution of theseand other psychological “assets.” Yet there isno complete way to discuss them with thelanguage of beliefs and desires, which is theonly language operating in quadrant I.

6.1 Can Neuroscience Save RationalChoice Economics?

Many neuroscientists are now using themost basic elements of rational choice theo-ry to explain what they see. Ironically, theyare taking up rational choice theory at thesame time as more and more economists aremoving away from rational choice toward abehavioral view anchored in limits on ration-ality, willpower, and greed (which we expectto be informed by neural detail). For exam-ple, neuroscientific studies of simple rewardcircuitry in rats and primates vindicate someof the simplest ideas of economics—namely,a “common brain currency” which permitssubstitution (Shizgal 1999), existence of neu-rons which encode expected reward (seeGlimcher 2002 and Figure 3), and revealed



preference (macaques viewing socially val-ued images, see Deaner, Khera, and Platt2005). Other groups are using Bayesianmodels. One study concluded that “the cen-tral nervous system therefore employs[Bayesian] probabilistic models during sen-sorimotor learning” (Konrad Körding andDaniel Wolpert 2004).

Our view is that establishing a neural basisfor some rational choice principles will notnecessarily vindicate the approach as widelyapplied to humans. The reason is that thestudies which have most clearly establishedexpected-value and Bayesian encoding usesimple tasks which monkeys and humans arewell-evolved to perform (e.g., reaching toearn a juice reward). It is quite possible thatsimple rational mechanisms are neurallyinstantiated to do these tasks well. But themost important kinds of economic behaviorinvolve manipulating abstract symbols, think-ing about groups of people and complexinstitutions, trading off very different types ofrewarding objects across time, and weighingthem by probabilities which are not alwayslearnable by experience. Ironically, rationalchoice models might therefore be most use-ful in thinking about the simplest kinds ofdecisions humans and other species make—involving perceptual tradeoffs, motor move-ments, foraging for food, and so forth—andprove least useful in thinking about abstract,complex, long-term tradeoffs which are thetraditional province of economic theory.

6.2 Incremental Versus RadicalNeuroeconomics

How should neuroeconomics contributeto advances in economic analysis? In theshort-run, an “incremental” approach inwhich psychological evidence suggests func-tional forms will help enhance the realism ofexisting models. For example, the two-parameter “ß–δ“ hyperbolic discountingapproach (where ß expresses preference forimmediacy, and is equal to one in the stan-dard model) is an example that has provenproductive theoretically (e.g., Laibson,

Repetto and Tobacman 1998; O’Donoghueand Rabin 1999). Laibson’s (2001) model ofhomeostatic response to environmental cuesin addiction is an incremental model wellgrounded in recent neuroscience.

However, we believe that in the long run amore “radical” departure from current theo-ry will become necessary, in the sense thatthe basic building blocks will not just consistof preferences, constrained optimization and(market or game-theoretic) equilibrium.After all, the point of constrained optimiza-tion is to model behavior precisely and pre-dict how behavior changes in response tochanges in budget constraints and prices.There is no reason other models startingfrom a very different basis could not be con-structed, while also predicting behavioralresponses to constraints and prices, and pre-dicting responses to other variables as well.Furthermore, thinking about the brain doesnot so much “falsify” rational choice theoriesas suggest entirely new distinctions andquestions.

For example, is learning that a particularperson cooperated with you in a gameencoded as a cognitive reputational statisticabout that person, like a numerical test score,or a “warm glow” which produces a surge ofdopamine when you see the person’s face(Singer et al. 2004; cf. James Rilling et al.2002). Standard game theory has no answerfor this question. But the distinction matters.If reputations are encoded dopaminergically,then they may spill over across groups ofpeople who look the same, for example, orwho are perceived as being part of a groupwho behave similarly (Bill McEvily et al.2003; Paul Healy 2004). Or the mechanismcan work in reverse—since attractive facesare known to produce dopaminergic surges(Aharon et al. 2001), the cortex may mistakethese positive signals of cooperativeness(which also produces pleasurable sensations)and automatically judge attractive faces aslikely to be cooperative. This simple neuralquestion about reward spillover might formthe foundation of group affiliation and social



26 Thanks to Nava Ashraf for clarifying this point.

capital and have something to do with whysome countries are rich and others poor.

Our main theme in this paper is that radi-cal models should respect the fact that brainmechanisms combine controlled and auto-matic processes, operating using cognitionand affect. The Platonic metaphor of reasonas a charioteer, driving twin horses of passionand appetite, is on the right track—exceptreason has its hands full with headstrongpassions and appetites.26 Of course, the chal-lenge in radical-style theorizing is to developmodels of how multiple mechanisms interactwhich are precise. Can this be done? Theanswer is Yes. Bernheim and Rangel (2004),Loewenstein and O’Donoghue (2004), andJess Benhabib and Alberto Bisin (2002) haveall proposed recent models with interactingmechanisms much like those in our Table 1.

Furthermore, while interactions of multi-ple brain mechanisms might appear to betoo radical a change from equilibrium withutility maximization, we think many familiartools can be used to do radical neuroeco-nomics. Interactions of cognition and affectmight resemble systems like supply anddemand, or feedback loops which exhibitmultiple equilibria. The interaction of con-trolled and automatic processes might belike an inventory policy or agency model inwhich a controller only steps in when anextreme state of the system (or unusualevent) requires controlled processes tooverride automatic ones. The influence ofaffect on choices is a very general type ofstate-dependence (where the “state” isaffective, and is influenced by external cuesand also by internal deliberation andrestraint). Instead of solving for equilibria inthese interacting-mechanism models, solvefor steady states or cyclic fluctuations.Instead of summarizing responses tochanges by comparative statics, studyimpulse-response functions.

Although we focused solely on applicationsof neuroscience to economics, intellectual

trade could also flow in the opposite direc-tion. Neuroscience is shot through withfamiliar economic language — delegation,division of labor, constraint, coordination,executive function — but these concepts arenot formalized in neuroscience as they are ineconomics. There is no overall theory of howthe brain allocates resources that are essen-tially fixed (e.g., blood flow and attention).An “economic model of the brain” could helphere. Simple economic concepts, like mech-anisms for rationing under scarcity, and gen-eral versus partial equilibrium responses toshocks, could help neuroscientists under-stand how the entire brain interacts. At atechnical level, neuroscientists are usingtools imported from econometrics, such asGranger causality (e.g., Wolfram Hesse et al.2003), to draw better inferences from neuraltime series. Finally, as the center of gravity inneuroscience research shifts from elemen-tary cognitive processes to the study of so-called higher functions—reasoning, socialinference, and decision making—neurosci-entists will increasingly reference, and drawinspiration from, the conceptual apparatus ofeconomics, a unique distillate of our century-long reflection on individual and strategicbehavior.

REFERENCES

Adolphs, Ralph, Daniel Tranel, Hanna Damasio, andAntonio R. Damasio. 1995. “Fear and the HumanAmygdala.” Journal of Neuroscience, 15(9): 5879–91.

Aharon, Itzhak, Nancy Etcoff, Dan Ariely, Chris F.Chabris, Ethan O’Connor, and Hans C. Breiter. 2001.“Beautiful Faces Have Variable Reward Value: fMRIand Behavioral Evidence.” Neuron, 32(3): 537–51.

Allman, John, Atiya Hakeem, and Karli Watson. 2002.“Two Phylogenetic Specializations in the HumanBrain.” Neuroscientist, 8(4): 335–46.

Ameriks, John, Andrew Caplin, and John Leahy. 2003.“Wealth Accumulation and the Propensity to Plan.”The Quarterly Journal of Economics, 118(3):1007–47.

Anderson, Adam K., Kalina Christoff, David Panitz,Eve De Rosa, and John D. Gabrieli. 2003. “NeuralCorrelates of the Automatic Processing of ThreatFacial Signals.” Journal of Neuroscience, 23(13):5627–33.

Ariely, Dan and George Loewenstein. 2003. “The Heatof the Moment: The Effect of Sexual Arousal onDecision Making,” CMU Working Paper.

Ariely, Dan, Uri Gneezy, George Loewenstein, and



Nina Mazar. 2004. “Large Stakes and Big Mistakes,”CMU Working Paper.

Armony, Jorge L., David Servan-Schreiber, JonathanD. Cohen, and Joseph E. LeDoux. 1995. “AnAnatomically Constrained Neural Network Model ofFear Conditioning.” Behavioral Neuroscience,109(2): 246–57.

Armony, Jorge L., David Servan-Schreiber, Jonathan D.Cohen, and Joseph E. LeDoux. 1997. “ComputationalModeling of Emotion: Explorations through theAnatomy and Physiology of Fear Conditioning.”Trends in Cognitive Sciences, 1(1): 28–34.

Barberis, Nicholas, Ming Huang, and Tano Santos.2001. “Prospect Theory and Asset Prices.” TheQuarterly Journal of Economics, 116(1): 1–53.

Bargh, John A. 1984. “Automatic and ConsciousProcessing of Social Information,” in Handbook ofSocial Cognition. R. S. Wyer Jr. and T. K. Srull, eds.Hillsdale, NJ: Erlbaum, 1–43.

Bargh, John A., Shelly Chaiken, Paula Raymond, andCharles Hymes. 1996. “The Automatic EvaluationEffect: Unconditional Automatic Attitude Activationwith a Pronunciation Task.” Journal of ExperimentalSocial Psychology, 32(1): 104–28.

Bargh, John A. and Tanya L. Chartrand. 1999. “TheUnbearable Automaticity of Being.” AmericanPsychologist, 54(7): 462–79.

Barlow, David H. 1988. Anxiety and its Disorders: TheNature and Treatment of Anxiety and Panic. NewYork: Guilford Press.

Baron-Cohen, Simon. 2000. “Theory of Mind andAutism: A Fifteen Year Review,” in UnderstandingOther Minds: Perspectives from DevelopmentalCognitive Neuroscience. Simon Baron-Cohen, HelenTager-Flusberg and Donald J. Cohen, eds. Oxford:Oxford University Press, 1–20.

Barraclough, Dominic J., Michelle L. Conroy, andDaeyeol Lee. 2004. “Prefrontal Cortex and DecisionMaking in a Mixed-Strategy Game.” NatureNeuroscience, 7(4): 404–10.

Baumeister, Roy F., Todd F. Heatherton, and DianneM. Tice. 1994. Losing Control: How and Why PeopleFail at Self-Regulation. San Diego: Academic Press.

Baumeister, Roy F. and Kathleen D. Vohs. 2003.“Willpower, Choice, and Self-control,” in Time andDecision: Economic and Psychological Perspectiveson Intertemporal Choice. George F. Loewenstein,Daniel Read and Roy F. Baumeister, eds. New York:Russell Sage Foundation, 13–86.

Bazerman, Max, George Loewenstein, and LeighThompson. 1989. “Social Utility and DecisionMaking in Interpersonal Contexts.” Journal ofPersonality and Social Psychology, 57(3): 426–41.

Bechara, Antoine, Antonio R. Damasio, HannaDamasio, and Steven W. Anderson. 1994.“Insensitivity to Future Consequences FollowingDamage to Human Prefrontal Cortex.” Cognition,50(1-3): 7–15.

Bechara, Antoine, Hanna Damasio, Antonio R.Damasio, and G. P. Lee. 1999. “DifferentContributions of the Human Amygdala andVentromedial Prefrontal Cortex to Decision-Making.” Journal of Neuroscience, 19(13): 5473–81.

Bechara, Antoine, Hanna Damasio, Daniel Tranel, and Antonio R. Damasio. 1997. “DecidingAdvantageously Before Knowing the AdvantageousStrategy.” Science, 275(5304): 1293–95.

Becker, Gary and Luis Rayo. 2004. “EvolutionaryEfficiency and Mean Reversion in Happiness.”University of Chicago Graduate School of Business.

Benartzi, Shlomo and Richard H. Thaler. 1995. “MyopicLoss Aversion and the Equity Premium Puzzle.” TheQuarterly Journal of Economics, 110(1): 73–92.

Benhabib, Jess and Alberto Bisin. 2002. “Self-Controland Consumption-Savings Decisions: CognitivePerspectives.” New York University.

Berkowitz, Leonard. 1999. “Anger,” in Handbook ofCognition and Emotion. Tim Dalgleish and Mick J.Power, eds. New York: Wiley, 411–28.

Bernheim, B. Douglas and Antonio Rangel. 2004.“Addiction and Cue-Conditioned CognitiveProcesses.” American Economic Review, 94(5):1558–90.

Berridge, Kent C. 1996. “Food Reward: BrainSubstrates of Wanting and Liking.” Neuroscienceand Biobehavioral Reviews, 20(1): 1–25.

Bertrand, Marianne and Sendhil Mullaninathan. 2004.“Are Emily and Greg More Employable than Lakishaand Jamal? A Field Experiment on Labor MarketDiscrimination,” NBER Working Paper 9873.

Bewley, Truman F. 2002. “Fairness, Reciprocity, andWage Rigidity,” Cowles Foundation Working Paper.http://cowles.econ.yale.edu/P/cd/d13b/d1383.pdf.

Bhatt, Meghana and Colin F. Camerer. In Press. “Self-Referential Strategic Thinking and Equilibrium asStates of Mind in Games: Evidence from fMRI.”Games and Economic Behavior.

Blair, R. James and Lisa Cipolotti. 2000. “ImpairedSocial Response Reversal: A Case of ‘AcquiredSociopathy.’” Brain, 123(6): 1122–41.

Bodner, Ronit and Drazen Prelec. 2003. “TheDiagnostic Value of Actions in a Self-SignalingModel,” in The Psychology of Economic DecisionsVol. 1. Isabelle Brocas and Juan D. Carillo, eds.Oxford: Oxford University Press.

Breiter, Hans C., I. Aharon, Daniel Kahneman, A. Dale,and Peter Shizgal. 2001. “Functional Imaging ofNeural Responses to Expectancy and Experience ofMonetary Gains and Losses.” Neuron, 30(2): 619–39.

Buck, Ross. 1999. “The Biological Affects: A Typology.”Psychological Review, 106(2): 301–36.

Cabanac, Michael. 1979. “Sensory Pleasure.”Quarterly Review of Biology, 54(1): 1–29.

Caggiula, Anthony R. and Bartley G. Hoebel. 1966.“‘Copulation–Reward Site’ in the PosteriorHypothalamus.” Science, 153(3741): 1284–85.

Camerer, Colin F. 2003. Behavioral Game Theory:Experiments in Strategic Interaction. RoundtableSeries in Behavioral Economics. Princeton: PrincetonUniversity Press; New York: Russell Sage Foundation.

Camerer, Colin F. and George Loewenstein. 2004.“Behavioral Economics: Past, Present, and Future,”in Advances in Behavioral Economics. Colin F.Camerer, George Loewenstein and Matthew Rabin,eds. Princeton: Princeton University Press.

Camerer, Colin F. and Teck Hua Ho. 1999.



“Experience-Weighted Attraction Learning inNormal Form Games.” Econometrica, 67(4): 827–74.

Camerer, Colin F., Teck Hua Ho, and Kuan Chong.2004. “A Cognitive Hierarchy Model of One-ShotGames.” Quarterly Journal of Economics, 119(3):861–98.

Camerer, Colin F., George Loewenstein, and MartinWeber. 1989. “The Curse of Knowledge in EconomicSettings: An Experimental Analysis.” Journal ofPolitical Economy, 97(5): 1232–54.

Camerer, Colin F. and Dan Lovallo. 1999.“Overconfidence and Excess Entry: AnExperimental Approach.” American EconomicReview, 89(1): 306–18.

Campfield, L. Arthur and Francoise J. Smith. 1990.“Systemic Factors in the Control of Food Intake,” inHandbook of Behavioural Neurobiology, Vol. 10:Neurobiology of Food and Fluid Intake. E. M.Stricker, ed. New York: Plenum, 183–206.

Caplin, Andrew and John Leahy. 2001. “PsychologicalExpected Utility Theory and Anticipatory Feelings.”The Quarterly Journal of Economics, 116(1): 55–79.

Carter, Rita. 1999. Mapping the Mind. Berkeley:University of California Press.

Chorvat, Terrence, Kevin McCabe and Vernon L.Smith. 2004. “Law and Neuroeconomics.” GeorgeMason Working Paper. http://papers.ssrn.com/sol13/papers.cfm?abstract_id=501063.

Cleckley, Hervey. 1941. The Mask of Sanity. St. Louis:C. V. Mosby.

Clore, Gerald L. 1992. “Cognitive Phenomenology:Feelings and the Construction of Judgment,” in TheConstruction of Social Judgments. Leonard L.Martin and Abraham Tesser, eds. Hillsdale:Erlbaum, 133–63.

Comings, David E. 1998. “The Molecular Genetics ofPathological Gambling.” CNS Spectrums, 3(6):20–37.

Cosmides, Leda. 1989. “The Logic of Social Exchange:Has Natural Selection Shaped How HumansReason? Studies with the Wason Selection Task.”Cognition, 31(3): 187–276.

Cosmides, Leda and John Tooby. 2004. “EvolutionaryPsychology and the Emotions,” in Handbook ofEmotions. Michael Lewis and Jeannette M.Haviland-Jones, eds. New York: Guilford.

Costa-Gomes, Miguel, Vincent P. Crawford, and BrunoBroseta. 2001. “Cognition and Behavior in Normal-Form Games: An Experimental Study.”Econometrica, 69(5): 1193–235.

Costa, Dora L. and Matthew E. Kahn. “Shame andOstracism: Union Army Deserters Leave Home.”MIT Working Paper. http://web.mit.edu/costa/www/shame11b.html.

Cottle, Thomas. 1974. The Present of Things Future:Explorations of Time in Human Experience. NewYork: Free Press.

Damasio, Antonio R. 1994. Descartes’ Error: Emotion,Reason, and the Human Brain. New York: G. P.Putnam.

Dana, Jason and George Lowenstein. 2003. “APsychological Perspective on the Influence of Giftsto Physicians from Industrry.” Journal of the

American Medical Association, 290(2): 252–55.de Becker, Gavin. 1997. The Gift of Fear: Survival

Signals That Protect Us from Violence. Boston:Little, Brown and Co.

Deaner, Robert O., Amit V. Khera, and Michael Platt.2005. “Monkeys Pay Per View: Adaptive Valuation ofSocial Images by Rhesus Macaques,” CurrentBiology. http://download.current-biology.com/pdfs/0960-9822/PIIS960982205001041.pdf

de Gelder, Beatrice, Jean Vroomen, Gilles Pourtois,and Lawrence Weiskrantz. 1999. “Non-ConsciousRecognition of Affect in the Absence of StriateCortex.” Neuroreport, 10(18): 3759–63.

De Houwer, Jan, Dirk Hermans, and Paul Eelen. 1998.“Affective and Identity Priming with EpisodicallyAssociated Stimuli.” Cognition and Emotion, 12(2):145–69.

Delgado, Mauricio R., Leigh E. Nystrom, C. Fissell, D.C. Noll, and Julie A. Fiez. 2000. “Tracking theHemodynamic Responses to Reward andPunishment in the Striatum.” Journal ofNeurophysiology, 84(6): 3072–77.

DeLillo, Don. White Noise. East Rutherford, NJ:Penguin Books.

della Vigna, Stefano and Ulrike Malmendier. 2003.“Overestimating Self-Control: Evidence from theHealth Club Industry,” Berkeley Working Paper.http://emlab.berkeley.edu/users/sdellavi/wp/self_control_nov03.pdf.

Denes-Raj, Veronika, Seymour Epstein, and JonathanCole. 1995. “The Generality of the Ratio–BiasPhenomenon.” Personality and Social PsychologyBulletin, 21(10): 1083–92.

Dubois, Samuel, Bruno Rossion, Christine Schiltz,Jean-Michel Bodart, Christian Michel, RaymondBruyer, and Marc Crommelinck. 1999. “Effect ofFamiliarity on the Processing of Human Faces.”Neuroimage, 9: 278–89.

Eisenberger, Naomi I., Matthew D. Lieberman, andKipling D. Williams. 2003. “Does Rejection Hurt?An fMRI Study of Social Exclusion.” Science,302(5643): 290–92.

Elbert, Thomas, Christo Pantev, Christian Wienbruch,Brigitte Rockstroh, and Edward Taub. 1995.“Increased Cortical Representation of the Fingers ofthe Left Hand in String Players.” Science, 270(5234):305–07.

Elster, Jon. 1977. Ulysses and the Sirens. Cambridge:Cambridge University Press.

Erk, Suzanne, Manfred Spitzer, Arthur P. Wunderlich,Lars Galley, and Henrik Walter. 2002. “CulturalObjects Modulate Reward Circuitry.” Neuroreport,13(18): 2499–503.

Farrer, Chlöé and Chris D. Frith. 2001. “ExperiencingOneself versus Another Person as Being the Cause ofan Action: The Neural Correlates of the Experienceof Agency.” Neuroimage, 15(3): 596–603.

Fazio, Russell, David M. Sanbonmatsu, Martha C.Powell, and Frank Kardes. 1986. “On the AutomaticActivation of Attitudes.” Journal of Personality andSocial Psychology, 50(2): 229–38.

Fehr, Ernst and Simon Gachter. 2000. “Cooperationand Punishment in Public Goods Experiments.”



American Economic Review, 90(4): 980–94.Fletcher, Paul C., Francesca Happe, Uta Frith, S. C.

Baker, Ray J. Dolan, Richard S. Frackowiak, andChris D. Frith. 1995. “Other Minds in the Brain: AFunctional Imaging Study of “Theory of Mind” inStory Comprehension.” Cognition, 57(2): 109–28.

Frederick, Shane and George Loewenstein. 2000. “ThePsychology of Sequence Preferences.” CMUWorking Paper.

Frederick, Shane, George Loewenstein, and TedO’Donoghue. 2002. “Time Discounting and TimePreference: A Critical Review.” Journal of EconomicLiterature, 40(2): 351–401.

Freeman, Walter J. and James W. Watts. 1942.Psychosurgery in the Treatment of Mental Disordersand Intractable Pain. Springfield: Thomas.

Fried, Itzhak. 1998. “Technical Comment: TheHippocampus and Human Navigation.” Science,282(5397): 2151.

Frijda, Nico. 1986. The Emotions. Cambridge:Cambridge University Press.

Frith, Uta. 2001a. “Mind Blindness and the Brain inAutism.” Neuron, 32(6): 969–79.

Frith, Uta. 2001b. “What Framework Should We Usefor Understanding Developmental Disorders?”Developmental Neuropsychology, 20(2): 555–63.

Fudenberg, Drew and David K. Levine. 1998. TheTheory of Learning in Games. Cambridge andLondon: MIT Press.

Gazzaniga, Michael S. and Joseph E. LeDoux. 1978.The Integrated Mind. New York: Plenum.

Gibbs, James, S. P. Madison, and Edmund T. Rolls.1981. “Satiety Role of the Small Intestine Examinedin Sham-Feeding Rhesus Monkeys.” Journal ofComparative Physiology and Psychology, 95(6):1003–15.

Gilbert, Daniel T. 2002. “Inferential Correction,” inHeuristics and Biases: The Psychology of IntuitiveJudgment. Thomas Gilovich, D. Griffin and DanielKahneman, eds. Cambridge: Cambridge UniversityPress, 167–84.

Gilbert, Daniel T. and Michael J. Gill. 2000. “TheMomentary Realist.” Psychological Science, 11(5):394–98.

Giner-Sorolla, Roger. 2001. “Guilty Pleasures andGrim Necessities: Affective Attitudes in Dilemmasof Self-Control.” Journal of Personality and SocialPsychology, 80(2): 206–21.

Giordano, Louis A., Warren K. Bickel, GeorgeLoewenstein, Eric A. Jacobs, Lisa Marsch, and Gary J.Badger. 2002. “Opioid Deprivation Affects HowOpioid-Dependent Outpatients Discount the Value ofDelayed Heroin and Money.” Psychopharmacology,163(2): 174–82.

Glimcher, Paul W. 2003. Decisions, Uncertainty, andthe Brain: The Science of Neuroeconomics.Cambridge and London: MIT Press.

Glimcher, Paul W., Michael Dorris, and Hannah Bayer.In Press. “Physiological Utility Theory and theNeuroeconomics of Choice.” Games and EconomicBehavior.

Gneezy, Uri and Jan Potters. 1997. “An Experiment onRisk Taking and Evaluation Periods.” The Quarterly

Journal of Economics, 112(2): 631–45.Gneezy, Uri and Aldo Rustichini. 2004. “Incentives,

Punishment and Behavior,” in Advances inBehavioral Economics. Colin F. Camerer, George F.Loewenstein and Matthew Rabin, eds. New York:Russell Sage Foundation, 572–89.

Gobet, Ferhand and Herbert A. Simon. 1996. “Recallof Random and Distorted Chess Positions:Implications for the Theory of Expertise.” Memoryand Cognition, 24(4): 493–503.

Goldman, Alvin. 2003. “Emotion, Mindreading,Simulation, and Modularity.” Other Minds: AnInterdisciplinary Conference. Institute of Cognitiveand Decision Sciences: University of Oregon,September 27–28, 2003.

Gollwitzer, Peter M., Kentaro Fujita, and GabrieleOettingen. 2004. “Planning and Implementation ofGoals,” in Handbook of Self-Regulation: Research,Theory, and Applications. Roy F. Baumeister andKathleen D. Vohs, eds. New York: Guilford Press,211–28.

Gonzalez, Roxanna and George Loewenstein. 2004.“Effects of Circadian Rhythm on Cooperation in an Experimental Game.” Working Paper. http://ssrn.com/abstract=485442.

Green, Leonard J. and Howard Rachlin. 1991.“Economic Substitutability of Electrical BrainStimulation, Food, and Water.” Journal of theExperimental Analysis of Behavior, 55(2): 133–43.

Greenberg, D., G. P. Smith, and J. Gibbs. 1990.“Intraduodenal Infusions of Fats Elicit Satiety inSham-Feeding Rats.” American Journal ofPhysiology: Regulatory, Integrative andComparative Physiology, 259(1 Pt. 2): R110–18.

Greenwald, Anthony G. 1992. “New Look 3:Unconscious Cognition Reclaimed.” AmericanPsychologist, 47(6): 766–79.

Greenwald, Anthony G., M. R. Klinger, and T. J. Liu.1989. “Unconscious Processing of DichopticallyMasked Words.” Memory and Cognition, 17(1):35–47.

Grether, David, Charles Plott, Daniel Rowe, MartinSereno, and John Allman. 2004. “Mental Processesand Strategic Equilibration: An fMRI Study ofSelling Strategy in Second Price Auctions.” CaltechWorking Paper No. 1189. http://www.hss.caltech.edu/SSpapers/wp1189.pdf.

Hadjikhani, Nouchine and Beatrice de Gelder. 2003.“Seeing Fearful Body Expressions Activates theFusiform Cortex and Amygdala.” Current Biology,13(24): 2201–05.

Haggard, Patrick, Sam Clark, and Jeri Kalogeras. 2002.“Voluntary Action and Conscious Awareness.”Nature Neuroscience, 5(4): 382–85.

Hahnloser, Richard H., Rahul Sarpeshkar, Misha A.Mahowald, Rodney J. Douglas, and H. SebastianSeung. 2000. “Digital Selection and AnalogueAmplification Coexist in a Cortex-Inspired SiliconCircuit.” Nature, 405(6789): 947–51.

Haier, Richard J., Benjamin V. Siegel, Jr., AndrewMacLachlan, Eric Soderling, Stephen Lottenberg,and Monte S. Buchsbaum. 1992. “Regional GlucoseMetabolic Changes After Learning a Complex



Visuospatial/Motor Task: a PET Study.” BrainResearch, 570(1–2): 134–43.

Happe, Francesca, Stefan Ehlers, Paul Fletcher, UtaFrith, Maria Johansson, Christopher Gillberg, RayDolan, Richard Frackowiak, and Chris Frith. 1996.“‘Theory of Mind’ in the Brain: Evidence from aPET Scan Study of Asperger Syndrome.”Neuroreport, 8(1): 197–201.

Hare, Robert D. 1965. “Psychopathy, Fear Arousal AndAnticipated Pain.” Psychological Reports, 16:499–502.

Hare, Robert D. 1966. “Temporal Gradient of FearArousal in Psychopaths.” Journal of AbnormalPsychology, 70(6): 442–45.

Hart, Allen J., Paul Whalen, Lisa Shin, SeanMcInerney, Hakan Fischer, and Scott Rauch. 2000.“Differential Response in the Human Amygdala toRacial Outgroup versus Ingroup Face Stimuli.”Neuroreport, 11(11): 2351–55.

Hastie, Reid. 1984. “Causes and Effects of CausalAttribution.” Journal of Personality and SocialPsychology, 46(1): 44–56.

Healy, Paul J. 2004. “Group Reputations andStereotypes as a Contract Enforcement Device.”Caltech Working Paper. http://kakutani.caltech.edu/~pj/papers/Healy-StereotypeContracts.pdf.

Hebb, Donald. 1949. The Organization of Behavior: ANeuropsychological Theory. New York: Wiley.

Hershey, John C., Howard C. Kunreuther, and Paul J.H. Schoemaker. 1982. “Sources of Bias inAssessment Procedures for Utility Functions.”Management Science, 28(8): 936–54.

Hesse, Wolfram, Eva Möller, Matthias Arnold, andBärbel Schack. 2003. “The Use of Time-VariantEEG Granger Causality for Inspecting DirectedInterdependencies of Neural Assemblies.” Journal ofNeuroscience Methods, 124(1): 27–44.

Hill, Elisabeth and David Sally. 2003. “Dilemmas andBargains: Autism, Theory-of-Mind, Cooperation andFairness.” University College London WorkingPaper. http://ssrn.com/abstract=407040.

Hirst, William. 1993. “On the Nature of Systems,” inConceptions of the Human Mind: Essays in Honor ofGeorge A. Miller. Gilbert Harman and George A.Miller eds. Hillsdale: Erlbaum, 1–11.

Hoch, Stephen J. and George F. Loewenstein. 1991.“Time-Inconsistent Preferences and Consumer Self-Control.” Journal of Consumer Research, 17(4):492–507.

Hoffman, Elizabeth, Kevin McCabe, Keith Shachat,and Vernon L. Smith. 1994. “Preferences, PropertyRights, and Anonymity in Bargaining Games.”Games and Economic Behavior, 7(3): 346–80.

Houston, David and Russell H. Fazio. 1989. “BiasedProcessing as a Function of Attitude Accessibility:Making Objective Judgments Subjectively.” SocialCognition, 7(1): 51-66.

Hsu, Ming, Meghana Bhatt, Ralph Adolphs, DanielTranel, and Colin Camerer. 2005. “Ambiguity-Aversion in the Brain: fMRI and Lesion-PatientEvidence.” Caltech Working Paper.

Hursh, Steven R. and B. H. Natelson. 1981. “ElectricalBrain Stimulation and Food Reinforcement

Dissociated by Demand Elasticity.” PhysiologicalBehavior, 26(3): 509–15.

Jevons, William S. 1871. The Theory of PoliticalEconomy. London: Macmillan and Co.

Johnson, Eric J., Colin F. Camerer, Sen Sankar, andTalia Tymon. 2002. “Detecting Failures of BackwardInduction: Monitoring Information Search inSequential Bargaining.” Journal of EconomicTheory, 104(1): 16–47.

Kahneman, Daniel. 2003. “Maps of BoundedRationality: Psychology for Behavioral Economics.”American Economic Review, 93(5): 1449–75.

Kahneman, Daniel and Shane Frederick. 2002.“Representativeness Revisited: AttributionSubstitution in Intuitive Judgment,” in Heuristics ofIntuitive Judgment: Extensions and Applications.Thomas Gilovich, Dale Griffin and DanielKahneman, eds. New York: Cambridge UniversityPress, 49–81.

Kirkpatrick, Lee A. and Seymour Epstein. 1992.“Cognitive–Experiential Self-Theory and SubjectiveProbability: Further Evidence for Two ConceptualSystems.” Journal of Personality and SocialPsychology, 63(4): 534–44.

Knutson, Brian and Richard Peterson. In Press.“Neurally Reconstructing Expected Utility.” Gamesand Economic Behavior.

Körding, Konrad P. and Daniel M. Wolpert. 2004.“Bayesian Integration in Sensorimotor Learning.”Nature, 427(6971): 244–47.

Kosslyn, Stephen M. 1994. Image and Brain.Cambridge: MIT Press.

Kruglanski, Arie W., Erik P. Thompson, E. ToryHiggins, M. Nadir Atash, Antonio Pierro, James Y.Shah, and Scott Spiegel. 2000. “To ‘Do the RightThing’ or to ‘Just Do It’: Locomotion and Assessmentas Distinct Self-Regulatory Imperatives.” Journal ofPersonality and Social Psychology, 79(5): 793–815.

Laibson, David. 2001. “A Cue-Theory of Consumption.”The Quarterly Journal of Economics, 116(1): 81–119.

Laibson, David I., Andrea Repetto, and JeremyTobacman. 1998. “Self-Control and Saving forRetirement.” Brookings Papers on EconomicActivity. 1: 91–196.

Lambrecht, Anja and Bernd Skiera. 2004. “Paying TooMuch and Being Happy About It: Causes andConsequences of Tariff Choice–Bias.” JohannWolfgang Goethe-University Frankfurt am MainWorking Paper.

Langer, Thomas and Craig R. Fox. “PartitionDependence and Naïve Diversification inInvestment Decision Making.” University ofMannheim Working Paper.

Le Bihan, Denis, Jean-François Mangin, Cyril Poupon,Chris A. Clark, Sabina Pappata, Nicolas Molko, andHughes Chabriat. 2001. “Diffusion Tensor Imaging:Concepts and Applications.” Journal of MagneticResonance Imaging, 13(4): 534–46.

Leboeuf, Robyn Aimee. 2002. “Alternating Selves andConflicting Choices: Identity Salience andPreference Inconsistency.” Dissertation AbstractsInternational, 63(2–B): 1088.

LeDoux, Joseph E. 1996. The Emotional Brain: The



Mysterious Underpinnings of Emotional Life. NewYork: Simon and Schuster.

Lee, Daeyeol, Michelle L. Conroy, Benjamin P.McGreevy and Dominic J. Barraclough. 2004.“Reinforcement Learning and Decision Making inMonkeys During a Competitive Game.” CognitiveBrain Research, 22(1): 45–58.

Lerner, Jennifer S. and Dacher Keltner. 2001. “Fear,Anger, and Risk.” Journal of Personality and SocialPsychology, 81(1): 146–59.

Libet, Benjamin. 1985. “Unconscious CerebralInitiative and the Role of Conscious Will in VoluntaryAction.” Behavior and Brain Sciences, 8(4): 529–66.

Lieberman, Matthew D. In Press. “Reflective andReflexive Judgment Processes: A Social CognitiveNeuroscience Approach,” in Responding to theSocial World: Implicit and Explicit Processes inSocial Judgments and Decisions. Joseph P. Forgas,Kipling D. Williams and William von Hippel, eds.Philadelphia: Psychology Press.

Lieberman, Matthew D., Ruth Gaunt, Daniel T.Gilbert, and Yaacov Trope. 2002. “Reflection andReflexion: A Social Cognitive NeuroscienceApproach to Attributional Interference,” inAdvances in Experimental Social Psychology. MarkP. Zanna, ed. New York: Academic Press, 199–249.

Lo, Andrew W. and Dmitry V. Repin. 2002. “ThePsychophysiology of Real-Time Financial RiskProcessing.” Journal of Cognitive Neuroscience,14(3): 323–39.

Loewenstein, George F. 1992. “The Fall and Rise ofPsychological Explanations in the Economics ofIntertemporal Choice,” in Choice Over Time.George Loewenstein and Jon Elster, eds. New York:Russell Sage Foundation, 3–34.

Loewenstein, George F. 1994. “The Psychology ofCuriosity: A Review and Reinterpretation.”Psychological Bulletin, 116(1): 75–98.

Loewenstein, George F. 1996. “Out of Control: VisceralInfluences on Behavior.” Organizational Behaviorand Human Decision Processes, 65(3): 272–92.

Loewenstein, George F. 1999. “Because It Is There:The Challenge of Mountaineering…for UtilityTheory.” Kyklos, 52(3): 315–43.

Loewenstein, George F. and Jennifer S. Lerner. 2003.“The Role of Affect in Decision Making,” inHandbook of Affective Sciences. Richard J. Davidson,Klaus R. Scherer and H. Hill Goldsmith, eds.Oxford: Oxford University Press, 619–42.

Loewenstein, George F. and Ted O’Donoghue. 2004.“Animal Spirits: Affective and Deliberative Influenceson Economic Behavior,” CMU Working Paper.

Loewenstein, George F. and Drazen Prelec. 1993.“Preferences for Sequences of Outcomes.”Psychological Review, 100(1): 91–108.

Loewenstein, George F., Leigh Thompson, and Max H.Bazerman. 1989. “Social Utility and DecisionMaking in Interpersonal Contexts.” Journal ofPersonality and Social Psychology, 57(3): 426–41.

Loewenstein, George F., Elke U. Weber, ChristopherK. Hsee, and Ned Welch. 2001. “Risk as Feelings.”Psychological Bulletin, 127(2): 267–86.

Loewenstein, George F., Robert Weber, Janine Flory,

Stephen Manuck and Matthew Muldoon. 2001.“Dimensions of Time Discounting.” Conference onSurvey Research on Household Expectations andPreferences. Ann Arbor, MI: November 2–3, 2001.

Logothetis, Nikos K., Jon Pauls, Mark Augath, TorstenTrinath, and Axel Oeltermann. 2001.“Neurophysiological Investigation of the Basis of thefMRI Signal.” Nature, 412(6843): 150–57.

Lucas, Robert E., Jr. 1978. “Asset Prices in anExchange Economy.” Econometrica, 46(6): 1429–45.

Lykken, David T. 1957. “A Study of Anxiety in theSociopathic Personality.” Journal of AbnormalPsychology, 55(1): 6–10.

MacLean, Paul D. 1990. The Triune Brain in Evolution:Role in Paleocerebral Function. New York: Plenum.

Maguire, Eleanor A., David G. Gadian, Ingrid S.Johnsrude, Catriona D. Good, John Ashburner,Richard S. Frackowiak, and Christopher D. Frith.2000. “Navigation-Related Structural Change in theHippocampi of Taxi Drivers.” Proceedings of theNational Academy of Sciences of the United States ofAmerica, 97(8): 4398–403.

Mailath, George. 1998. “Do People Play NashEquilibrium? Lessons from Evolutionary GameTheory.” Journal of Economic Literature, 36(3):1347–74.

Mailath, George J., Michihiro Kandori, and RafaelRob. 1993. “Learning, Mutation, and Long RunEquilibria in Games.” Econometrica, 61(1): 29–56.

Mandler, George. 1982. “The Structure of Value:Accounting for Taste,” in Affect and Cognition: TheSeventeenth Annual Carnegie Symposium onCognition. M. S. Clark and S. T. Fiske, eds. Hillsdale:Erlbaum, 3–36.

Manuck, Stephen B., Janine Flory, Matthew Muldoon,and Robert E. Ferrell. 2003. “A Neurobiology ofIntertemporal Choice,” in Time and Decision:Economic and Psychological Perspectives onIntertemporal Choice. George F. Loewenstein,Daniel Read and Roy F. Baumeister, eds. New York:Russell Sage Foundation, 139–72.

Marshall, J. F., J. S. Richardson, and P. Teitelbaum.1974. “Nigrostriatal Bundle Damage and the LateralHypothalamic Syndrome.” Journal of Comparativeand Physiological Psychology, 87(5): 808–30.

McCabe, Kevin, Daniel Houser, Lee Ryan, Vernon L.Smith, and Theodore Trouard. 2001. “A FunctionalImaging Study of Cooperation in Two-PersonReciprocal Exchange.” Proceedings of the NationalAcademy of Sciences of the United States of America,98(20): 11832–35.

McClure, Samuel M., David I. Laibson, George F.Loewenstein and Jonathan D. Cohen. 2004.“Separate Neural Systems Value Immediate andDelayed Monetary Rewards.” Science, 306(5695):503–07.

McConnell, Allen and Jill Leibold. 2001. “RelationsAmong the Implicit Association Test, DiscriminatoryBehavior, and Explicit Measures of Racial Attitudes.”Journal of Experimental Social Psychology, 37(5):435–42.

McElroy, Susan L., Andrew Satlin, Harrison Pope,Paul E. Keck, and James I. Hudson. 1991.



“Treatment of Compulsive Shopping withAntidepressants: A Report of Three Cases.” Annalsof Clinical Psychiatry, 3(3): 199–204.

McEvily, Bill, Roberto Weber, Cristina Bicchieri andViolet Ho. 2003. “Can Groups Be Trusted? AnExperimental Study of Collective Trust,” CMUWorking Paper. http://www.andrew.cmu/user/rwe-ber/CollectiveTrust10.pdf.

McNeil, Barbara J., Stephen G. Pauker, Harold C. Sox,Jr., and Amos Tversky. 1982. “On the Elicitation ofPreferences for Alternative Therapies.” NewEngland Journal of Medicine, 306(21): 1259–62.

Medin, Douglas L. and Max H. Bazerman. 1999.“Broadening Behavioral Decision Research:Multiple Levels of Cognitive Processing.”Psychonomic Bulletin and Review, 6(4): 533–47.

Mendelson, Joseph. 1967. “Lateral HypothalamicStimulation in Satiated Rats: The Rewarding Effectsof Self-Induced Drinking.” Science, 157(3792):1077–79.

Metcalfe, Janet and Walter Mischel. 1999. “A Hot/Cool-System Analysis of Delay of Gratification: Dynamicsof Willpower.” Psychological Review, 106(1): 3–19.

Miller, George A., Eugene Galanter, and Karl H.Pribram. 1960. Plans and the Structure of Behavior.New York: Holt.

Mischel, Walter, Ebbe B. Ebbesen, and Antonette R.Zeiss. 1972. “Cognitive and Attentional Mechanismsin Delay of Gratification.” Journal of Personality andSocial Psychology, 21(2): 204–18.

Mischel, Walter and Robert Metzner. 1962. “Preferencefor Delayed Reward as a Function of Age,Intelligence, and Length of Delay Interval.” Journalof Abnormal Social Psychology, 64(6): 425–31.

Mobbs, Dean, Michael D. Greicius, Eiman Abdel-Azim, Vinod Menon, and Allan L. Reiss. 2003.“Humor Modulates the Mesolimbic RewardCenters.” Neuron, 40(5): 1041–48.

Montague, P. Read and Gregory S. Berns. 2002.“Neural Economics and the Biological Substrates ofValuation.” Neuron, 36(2): 265–84.

Montague, P. Read, Gregory S. Berns, Jonathan D.Cohen, Samuel M. McClure, Giuseppe Pagnoni,Mukesh Dhamala, Michael C. Wiest, Igor Karpov,Richard D. King, Nathan Apple, and Ronald E.Fisher. 2002. “Hyperscanning: Simultaneous fMRIDuring Linked Social Interactions.” Neuroimage,16(4): 1159–64.

Moreyra, Paula, Angela A. Ibáñez, Jerónimo Saiz-Ruiz,Kore Nissenson, and Carlos Blanco. 2000. “Reviewof the Phenomenology, Etiology and Treatment ofPathological Gambling.” German Journal ofPsychiatry, 3(2): 37–52.

Morris, John S., C. Buchel, and Raymond J. Dolan.2001. “Parallel Neural Responses in AmygdalaSubregions and Sensory Cortex During Implicit FearConditioning.” Neuroimage, 13(6 Pt. 1): 1044–52.

Mullainathan, Sendhil. 2002. “Thinking ThroughCategories,” MIT Working Paper.

National Academy of Sciences. 1999. PathologicalGambling. Washington, D.C.: National Academy Press.

Nichols, M. James and William T. Newsome. 2002.“Middle Temporal Visual Area Microstimulation

Influences Veridical Judgments of MotionDirection.” Journal of Neuroscience, 22(21): 9530–40.

Nisbett, Richard E. and Timothy D. Wilson. 1977.“Telling More Than We Can Know: Verbal Reportson Mental Processes.” Psychological Bulletin, 84(3):231–59.

O’Donoghue, Ted and Matthew Rabin. 1997.“Addiction and Self-Control,” in Addiction: Entriesand Exits. Jon Elster, ed. New York: Russell SageFoundation, 1999, 169–206.

O’Donoghue, Ted and Matthew Rabin. 1999. “Doing ItNow or Later.” American Economic Review, 89(1):103–24.

Ochsner, Kevin N., Silvia A. Bunge, James J. Gross, andJames D.E. Gabrieli. 2002. “Rethinking Feelings: AnfMRI Study of the Cognitive Regulation of Emotion.”Journal of Cognitive Neuroscience, 14(8): 1215–29.

Ochsner, Kevin N. and Matthew D. Lieberman. 2001.“The Emergence of Social Cognitive Neuroscience.”American Psychologist, 56(9): 717–34.

Olds, James and Peter Milner. 1954. “PositiveReinforcement Produced by Electrical Stimulationof Septal Area and Other Regions of Rat Brain.”Journal of Comparative and PhysiologicalPsychology, 47(6): 419–27.

Oswald, Andrew J. 1997. “Happiness and EconomicPerformance.” Economic Journal, 107(445): 1815–31.

Panksepp, Jaak. 1998. Affective Neuroscience. Oxford:Oxford University Press.

Parsons, Lawrence M. and Daniel Osherson. 2001.“New Evidence for Distinct Right and Left BrainSystems for Deductive versus ProbabilisticReasoning.” Cerebral Cortex, 11(10): 954–65.

Persico, Nicola, Andrew Postlewaite, and DanSilverman. “The Effect of Adolescent Experience onLabor Market Outcomes: The Case of Height,”University of Pennsylvania Working Paper.

Persinger, Michael A. and Faye Healey. 2002.“Experimental Facilitation of the Sensed Presence:Possible Intercalation Between the HemispheresInduced by Complex Magnetic Fields.” Journal ofNervous and Mental Disease, 190(8): 533–41.

Peters, Ellen and Paul Slovic. 2000. “The Springs ofAction: Affective and Analytical InformationProcessing in Choice.” Personality and SocialPsychology Bulletin, 26(12): 1465–75.

Phelps, Elizabeth, Kevin J. O’Connor, William A.Cunningham, E. Sumie Funayama, J. ChristopherGatenby, John C. Gore, and Mahzarin R. Banaji.2000. “Performance on Indirect Measures of RaceEvaluation Predicts Amygdala Activity.” Journal ofCognitive Neuroscience, 12(5): 729–38.

Pillutla, Madan M. and Xiao-Ping Chen. 1999. “SocialNorms and Cooperation in Social Dilemmas: TheEffects of Context and Feedback.” OrganizationalBehavior and Human Decision Processes, 78(2):81–103.

Platt, Michael L. and Paul W. Glimcher. 1999. “NeuralCorrelates of Decision Variables in Parietal Cortex.”Nature, 400(6741): 233–38.

Prelec, Drazen and George F. Loewenstein. 1998.“The Red and the Black: Mental Accounting ofSavings and Debt.” Marketing Science, 17(1): 4–28.



Prelec, Drazen and Duncan Simester. 2001. “AlwaysLeave Home Without It.” Marketing Letters, 12(1):5–12.

Quattrone, George A. and Amos Tversky. 1984.“Causal Versus Diagnostic Contingencies: On Self-Deception and on the Voter’s Illusion.” Journal ofPersonality and Social Psychology, 46(2): 237–48.

Rabin, Matthew. 1993. “Incorporating Fairness intoGame Theory and Economics.” American EconomicReview, 83(5): 1281–302.

Read, Daniel and George Loewenstein. 1995. “TheDiversification Bias: Explaining the DifferenceBetween Prospective and Real-Time Taste forVariety.” Journal of Experimental Psychology:Applied, 1(1): 34–49.

Read, Daniel, George Loewenstein, and ShobanaKalyanaraman. 1999. “Mixing Virtue and Vice: TheCombined Effects of Hyperbolic Discounting andDiversification.” Journal of Behavioral DecisionMaking, 12(4): 257–73.

Rilling, James K., D. A. Gutman, T. R. Zeh, GiuseppePagnoni, Greg S. Berns, and Clint D. Kilts. 2002. “ANeural Basis for Social Cooperation.” Neuron, 35(2):395–405.

Rolls, Edmund T. 1999. The Brain and Emotion. NewYork: Oxford University Press.

Romer, Paul M. 2000. “Thinking and Feeling.”American Economic Review, 90(2): 439–43.

Ross, Lee, Mark R. Lepper, and Michael Hubbard.1975. “Perseverance in Self-Perception and SocialPerception: Biased Attributional Processes in theDebriefing Paradigm.” Journal of Personality andSocial Psychology, 32(5): 880–92.

Rowe, Andrea D., Peter R. Bullock, Charles E. Polkey,and Robin G. Morris. 2001. “ ‘Theory of Mind’Impairments and their Relationship to ExecutiveFunctioning Following Frontal Lobe Excisions.”Brain, 124(3): 600–16.

Rumelhart, David E. and James L. McClelland. 1986.Parallel Distributed Processing: Explorations in theMicrostructure of Cognition, Vol. 1: Foundations.Cambridge: MIT Press.

Rustichini, Aldo, John Dickhaut, Paolo Ghirardato, KipSmith and Jose V. Pardo. 2002. “A Brain ImagingStudy of Procedural Choice,” University ofMinnesota Working Paper.

Samuelson, Paul. 1937. “A Note on Measurement ofUtility.” Review of Economic Studies, 4: 155–61.

Sanfey, Alan G., James K. Rilling, Jessica A. Aaronson,Leigh E. Nystrom, and Jonathan D. Cohen. 2003.“The Neural Basis of Economic Decision-Making inthe Ultimatum Game.” Science, 300(5626): 1755–58.

Saxe, Rebecca and Nancy Kanwisher. 2003. “PeopleThinking about Thinking People: The Role of theTemporo–Parietal Junction in ‘Theory of Mind.’”Neuroimage, 19(4): 1835–42.

Schelling, Thomas C. 1978. “Egonomics, or the Art ofSelf-Management.” American Economic Review,68(2): 290–94.

Schelling, Thomas C. 1984. “The Intimate Contest forSelf-Command,” in Choice and Consequence.Cambridge: Harvard University Press, 57–81.

Schneider, Walter and Richard M. Shiffrin. 1977.

“Controlled and Automatic Human InformationProcessing: I. Detection, Search and Attention.”Psychological Review, 84(1): 1–66.

Schultz, Wolfram. 2002. “Getting Formal withDopamine and Reward.” Neuron, 36(2): 241–63.

Schultz, Wolfram and Anthony Dickinson. 2000.“Neuronal Coding of Prediction Errors.” AnnualReview of Neuroscience, 23: 473–500.

Schwarz, Norbert. 1990. “Feelings as Information:Informational and Motivational Functions ofAffective States,” in Handbook of Motivation andCognition: Foundations of Social Behavior. E. T.Higgins and R. M. Sorrentino, eds. New York:Guilford Press, 527–61.

Schwarz, Norbert and Gerald L. Clore. 1983. “Mood,Misattribution and Judgments of Well-Being:Informative and Directive Functions of AffectiveStates.” Journal of Personality and SocialPsychology, 45(3): 513–23.

Shallice, Timothy and Paul W. Burgess. 1996. “TheDomain of Supervisory Processes and TemporalOrganization of Behaviour.” PhilosophicalTransactions of the Royal Society B: BiologicalSciences, 351(1346): 1405–12.

Shapiro, David. 1965. Neurotic Styles. New York: BasicBooks.

Shefrin, Hersh and Richard Thaler. 1988. “TheBehavioral Life Cycle Hypothesis.” EconomicInquiry, 26(4): 609–43.

Sherry, David F. and Daniel L. Schacter. 1987. “TheEvolution of Multiple Memory Systems.”Psychological Review, 84(2): 127–90.

Shiffrin, Richard M. and Walter Schneider. 1977.“Controlled and Automatic Human InformationProcessing: II. Perceptual Learning, AutomaticAttending and a General Theory.” PsychologicalReview, 84(2): 127–90.

Shiv, Baba and Alexander Fedorikhin. 1999. “Heartand Mind in Conflict: The Interplay of Affect andCognition in Consumer Decision Making.” Journalof Consumer Research, 26(3): 278–92.

Shiv, Baba, George Loewenstein, Antoine Bechara,Antonio Damasio and Hanna Damasio. 2005.“Investment Behavior and the Dark Side ofEmotion.” Psychological Science. Forthcoming.

Shizgal, Peter. 1999. “On the Neural Computation ofUtility: Implications from Studies of BrainReward,” in Foundations of Hedonic Psychology:Scientific Perspectives on Enjoyment and Suffering.Daniel Kahneman, Edward Diener and NorbertSchwarz eds. New York: Russell Sage Foundation,502–26.

Simon, Herbert A. 1967. “Motivational and EmotionalControls of Cognition.” Psychological Review, 74(1):29–39.

Simonson, Itamar. 1990. “The Effect of PurchaseQuantity and Timing on Variety-Seeking Behavior.”Journal of Marketing Research, 27(2): 150–62.

Singer, Tania, Stefan J. Kiebel, Joel S. Winston,Raymond J. Dolan, and Chris D. Frith. 2004. “BrainResponses to the Acquired Moral Status of Faces.”Neuron, 41(4): 653–62.

Sloman, Steven A. 1996. “The Empirical Case for Two



Systems of Reasoning.” Psychological Bulletin,119(1): 3–22.

Smith, Trenton. “The McDonald’s Equilibrium:Advertising, Empty Calories, and the EndogenousDetermination of Dietary Preferences,” Universityof Bonn Working Paper. http://www.international.ucla.edu/cms/files/smithnov13.pdf.

Strack, Fritz, Leonard L. Martin, and Sabine Stepper.1988. “Inhibiting and Facilitating Conditions of theHuman Smile: A Nonobtrusive Test of the FacialFeedback Hypothesis.” Journal of Personality andSocial Psychology, 54(5): 768–77.

Titmuss, Richard M. 1970. The Gift Relationship.London: Allen and Unwin.

Train, Kenneth E. 1991. Optimal Regulation: TheEconomic Theory of Natural Monopoly. Cambridge:MIT Press.

Train, Kenneth E., Daniel L. McFadden, and MosheBen-Akiva. 1987. “The Demand for Local TelephoneService: A Fully Discrete Model of ResidentialCalling Patterns and Service Choices.” Rand Journalof Economics, 18(1): 109–23.

Tversky, Amos and Daniel Kahneman. 1983.“Extensional Versus Intuitive Reasoning: TheConjunction Fallacy in Probability Judgment.”Psychological Review, 90(4): 293–315.

Van Boven, Leaf, George Loewenstein, and DavidDunning. 2003. “Mispredicting the EndowmentEffect: Underestimation of Owners’ Selling Prices byBuyer’s Agents.” Journal of Economic Behavior andOrganization, 51(3): 351–65.

Viner, Jacob. 1925. “The Utility Concept in ValueTheory and its Critics.” Journal of Political Economy,33(4): 369–87.

von Melchner, Laurie, Sarah L. Pallas, and MrigankaSur. 2000. “Visual Behaviour Mediated by RetinalProjections Directed to the Auditory Pathway.”Nature, 404(6780): 871–76.

Vuilleumier, Patrik, Jorge L. Armony, Jon Driver, andRay J. Dolan. 2001. “Effects of Attention andEmotion on Face Processing in the Human Brain: AnEvent-Related fMRI Study.” Neuron, 30(3): 829–41.

Wegner, Daniel M. and Thalia Wheatley. 1999.“Apparent Mental Causation. Sources of theExperience of Will.” American Psychologist, 54(7):480–92.

Wells, Gary L. and Richard E. Petty. 1980. “TheEffects of Overt Head Movements on Persuasion:Compatibility and Incompatibility of Resources.”Basic and Applied Social Psychology. 1(3): 219–30.

Whalen, Paul J., S. L. Rauch, Nancy Etcoff, S. C.McInerney, M. Lee, and M. A. Jenike. 1998.“Masked Presentations of Emotional FacialExpressions Modulate Amygdala Activity withoutExplicit Knowledge.” Journal of Neuroscience, 18(1):411–18.

Whaling, Carol S., M. M. Solis, Allison J. Doupe, Jill A.Soha, and Peter Marler. 1997. “Acoustic and Neural

Bases for Innate Recognition of Song.” Proceedingsof the National Academy of Sciences of the UnitedStates of America, 94(23): 12694–98.

Wheeler, Mary E. and Susan T. Fiske. 2005.“Controlling Racial Prejudice: Social–CognitiveGoals Affect Amygdala and Stereotype Activation.”Psychological Science, 16(1): 56–63.

Wilson, Timothy D., Samuel Lindsey, and Tonya Y.Schooler. 2000. “A Model of Dual Attitudes.”Psychological Review, 107(1): 101–26.

Wilson, Timothy D., Douglas J. Lisle, Jonathan W.Schooler, Sara D. Hodges, Kristen J. Klaaren, andSuzanne J. LaFleur. 1993. “Introspecting aboutReasons Can Reduce Post-Choice Satisfaction.”Personality and Social Psychology Bulletin, 19(3):331–39.

Wilson, Timothy D. and Jonathan W. Schooler. 1991.“Thinking Too Much: Introspection Can Reduce theQuality of Preferences and Decisions.” Journal ofPersonality and Social Psychology, 60(2): 181–92.

Windschitl, Paul D. and Gary L. Wells. 1998. “TheAlternative–Outcomes Effect.” Journal ofPersonality and Social Psychology, 75(6): 1411–23.

Winkielman, Piotr and Kent C. Berridge. 2004.“Unconscious Emotion.” Current Directions inPsychological Science, 13(3): 120–23.

Wise, Roy A. 1996. “Addictive Drugs and BrainStimulation Reward.” Annual Review ofNeuroscience, 19: 319–40.

Wolford, George, Michael B. Miller, and MichaelGazzaniga. 2000. “The Left Hemisphere’s Role inHypothesis Formation.” Journal of Neuroscience,20(6): 1–4.

Zajonc, Robert B. 1980. “Feeling and Thinking:Preferences Need No Inferences.” AmericanPsychologist, 35(2): 151–75.

Zajonc, Robert B. 1984. “On the Primacy of Affect.”American Psychologist, 39(2): 117–23.

Zajonc, Robert B. 1998. “Emotions,” in Handbook ofSocial Psychology. Daniel T. Gilbert, Susan T. Fiskeand Gardner Lindzey, eds. New York: OxfordUniversity Press, 591–632.

Zajonc, Robert B. and Daniel N. McIntosh. 1992.“Emotions Research: Some Promising Questionsand Some Questionable Promises.” PsychologicalScience, 3(1): 70–74.

Zak, Paul. In Press. “Neuroeconomics.” Transactions ofthe Royal Philosophical Society B.

Zak, Paul, Robert Kurzban and William Matzner. 2003.“Oxytocin is Associated with Interpersonal Trust inHumans.” Under Review.

Zamir, Shmuel. 2000. “Rationality and Emotions inUltimatum Bargaining,” Discussion Paper #222.http://www.ma.huji.ac.il/~zamir/dp222.pdf.

Zink, Caroline F., Giuseppe Pagnoni, M. E. Martin, M.Dhamala, and Greg S. Berns. 2003. “Human StriatalResponse to Salient Nonrewarding Stimuli.” Journalof Neuroscience, 23(22): 8092–97.


Neuroeconomics How Neuroscience Can Inform Economics

Documents

Transcript of Neuroeconomics How Neuroscience Can Inform Economics