Neuroeconomics

Post-publication activity

Dr. Paul W. Glimcher, Center for Neural Science, New York University, NY

Over the course of the last three centuries, both social scientists and natural scientists have tried to understand how we make decisions, but using entirely different strategies. Since the late 1990s a group of interdisciplinary scholars have begun to combine the social and natural scientific approaches to the study of choice into an emerging synthetic discipline now called Neuroeconomics. The central assumption of this discipline is that by combining both theoretical and empirical tools from neuroscience, psychology and economics into a single approach, the resulting synthesis will provide insights valuable to all three parent disciplines. Studies conducted to date seem to support that conclusion. Theories from economics and psychology have already begun to restructure our neurobiological understanding of decision making, and a number of recent neurobiological findings are beginning to suggest constraints on theoretical models of choice developed in both economic and psychological domains.

Traditional Studies of Decision-Making

Since ancient times decision making has been studied by scholars at many levels of reduction, but in general the study of choice has been partitioned into three main approaches.

For most economists the goal of studying human choice behavior is prediction. These social scientists seek to develop formal mathematical models, typically based on a rigorous axiomatic foundation, that can predict the choices humans do, or should, make. These models typically take as inputs the state of the external world and generate as outputs the actual choices made by human choosers. For a mainstream economist, a model is useful if it makes accurate predictions; whether or not the algorithm it employs mimics the actual process of decision making is irrelevant to accomplishing this end. For this reason, economic studies of decision making can be viewed as aimed towards achieving both the most compact and the most abstract models of choice possible. The products of economics are high-level, and often normative, theories that state testable behavioral hypotheses.

At a lower level of reduction, psychologists studying the mechanisms of judgment and decision seek to understand the mental constructs that guide decision making at a more process-based level of analysis. Mental processes like the fear of losses or the human tendency to overestimate low probabilities form the algorithmic components of psychological models of choice. These models seek not just to predict behavior but to capture accurately the mental events that precede choice. As such, they are much more complicated that economic models. Although this mental complexity often makes them more realistic it does so at a cost, because these models are so complicated they can often be hard to test completely.

At a yet lower level of reduction neurobiologists have been trying to understand the neural pathways and computations that give rise to decision making behavior. These natural scientists have sought to understand, at a physical level, how it is that the brain achieves choice by studying the computational architecture of the brain. Of course the challenge neuroscientists face is one of scale. Understanding how decisions are made simply by tracing neural pathways has constrained neurobiologists to studying only very simple decisions, choices that an economist or psychologist would consider uninteresting.

The goal of Neuroeconomics is to combine these three approaches into a single discipline that employs constraints and insights from each level of analysis to inspire and constrain insights at other levels of analysis. In the future it may even be the case that Neuroeconomics may come to encompass other areas that also engage the study of decision; areas like control theory and basic statistics. In practice, however, this combination of disciplines has been largely restricted to combining neuroscience, psychology and economics and has proceeded as a two stage process. In the first stage, high level theories from the social sciences have been used to organize more biological insights into a unified conceptual framework. Economic theories like Expected Utility or Behavioral Game Theory have been used to define the computational goal of the neural architecture for decision making, a definition that both constrains what kinds of measurements should be made by neurobiologists and provides a theoretical framework for organizing the huge amount of neurobiological data on choice already available. In the second stage, neurobiological insights into the algorithms by which choice is accomplished can be expected to serve as constraints on high-level theories. If, for example, it can be shown that the neural architecture cannot perform some class of computation, then Neuroeconomists take this as evidence that high level theories which rely on that computation do not merit further inquiry.

In fairness it must be stated that to date only the first of these two stages has been widely accepted as productive and valuable by all of Neuroeconomics’ parent disciplines. While nearly all neurobiologists recognize the impact economic theory has had on biological studies of choice, many mainstream economists doubt that the second stage of this process will (or even should) occur. These social scientists doubt, in a general way, that the reductive program of the natural sciences can be extended to the social sciences. This is a point to which we return later in this article.

Some Influential Events in Neuroeconomics

Probably the first paper to explicitly combine neuroscientific data and a rigorous mathematical theory from the social sciences was Peter Shizgal and Kent Conover’s 1996 review in Current Directions in Psychological Science: “On the neural computation of Utility”. The paper sought to describe the neurobiological substrate for choice in rats using a normative economic theory. In 1999 this was followed by Michael Platt and Paul Glimcher’s publication of “Neural correlates of decision variables in parietal cortex” which argued that: “Neurobiologists have begun to focus increasingly on the study of sensory-motor processing, but many of the models used to describe these processes remain rooted in the classic reflex” and went on to “describe a formal economic-mathematical approach for the physiological study of the sensory-motor process, or decision-making”. Empirically, the paper demonstrated that the activity of individual neurons in the posterior parietal cortex encoded both the probability and magnitude of reward as would be predicted by most economic theories if these neurons participated in decision-making. Within neurobiological circles this paper, which sought to use economic approaches to studying choice in monkeys, was rapidly followed by a suite of papers in both humans and other animals uniting both economic and psychological theories of choice with measurements in human and animal brains.

The first of these neuroeconomic studies in humans were a pair of papers published in 2001. The first of these papers appeared in the Journal Neuron (Breiter et al., 2001) and reflected a collaboration between the fMRI pioneer Hans Breiter, Shizgal, and the Princeton psychologist/economist Daniel Kahneman (who would win the Nobel prize for his contribution to behavioral economics the following year). That paper employed the psychological Prospect theory of choice developed by Kahneman and his collaborator Amos Tversky, this time in a brain scanning experiment. In that paper, the perceived desirability of a particular outcome in a lottery the subject was asked to play (in this case winning zero dollars in a ‘wheel of fortune’ game) was manipulated by changing the values of two other possible lottery outcomes. When winning zero dollars is the worst of three possible outcomes Kahneman and Tversky’s (1979) Prospect theory predicts that subjects should view it negatively, but when it is the best of the three outcomes then subjects should view it more positively. The scanning experiment revealed that brain activation in the ventral striatum matched these predicted subjective valuations.

The second of these papers reflected a collaboration between the economist Kevin McCabe, his colleague Vernon Smith (who would share the Nobel prize with Kahneman the following year for his contributions to experimental economics) and a team that included economists, a psychologist and a biomedical engineer (McCabe et al., 2001). This also represented the first use of game theory in a human neurobiological experiment. In that paper, subjects played a trust game either against an anonymous human opponent or against a computer. The neurobiological data revealed that in some subjects the medial prefrontal cortex is more active when subjects play a cooperative strategy than when they show a lack of trust in their game theoretic opponent.

Perhaps the critical insight that these first three papers provided was evidence that the decision-making systems of the brain can be viewed as a fundamentally two-part system. Areas in the frontal cortex and basal ganglia form the first of these two parts. We now know that these areas learn and compute the values of available actions and it is as a set of valuation structures that these areas principally contribute to decision-making. The outputs of these structures then appear to be passed to fronto-parietal circuits which actually ‘decide’ between options based on these antecedent valuations and pass these decisions on to the motor system for execution. Subsequent studies have largely supported this segregation of the neural architecture into valuation and choice systems, although the levels of interconnection between these two systems are only now beginning to be explored.

Another major advance in the history of neuroeconomics was presented in 2005 by Michael Kosfeld, Ernst Fehr and their colleagues. This paper was the first demonstration of a neuropharmacological manipulation that alters behavior in a manner that can be interpreted with regard to normative theory. In that paper, subjects were asked to play a trust game much like the one examined by McCabe and colleagues. The experimenters’ critical manipulation was to increase brain levels of the neurotransmitter oxytocin before some of the players made their decisions about whether or not to trust their opponents. They found that players in the game were generally more trusting if they had received oxytocin than if they had been treated with a control substance. What was most interesting about this study from a neuroeconomic point of view was the demonstration that administration of this endogenously produced hormone altered the choice behavior of subjects in a way that went against an existing and well described normative economic theory.

Challenges to Neuroeconomics

During this initial period a set of summary reviews began to emerge that served as manifestos for the emerging Neuroeconomic discipline. In 2003 Glimcher published a book, directed primarily at neuroscientists, that reviewed the history of neuroscience and argued that this history was striking in its lack of normative models for higher cognitive function. Glimcher proposed that economics could serve as the source for this much needed normative theory. Shortly thereafter Camerer, Loewenstein and Prelec published a paper under the title of Neuroeconomics: How neuroscience can inform economics (2005) which also served as a manifesto, but this time from the economic side.

In that paper, Camerer and colleagues argued that the many failures of traditional economics to make accurate predictions about human behavior reflected an inattention to mechanism. Understanding how decisions are made by the brain, they proposed, would yield algorithmic alternatives to neoclassical theory with enhanced predictive power. In the paper itself, the authors provided examples of neuroeconomic studies that should in the future provide such algorithmic insights and should thus either constrain or direct future studies in economics. Although the logic of this proposal is clear, it is admittedly true that there have been very few insights from neurobiological studies that constrain economic theory. Noting that, Princeton’s Faruk Gul and Wolfgang Pesendorfer published in 2008 a widely read attack on Neuroeconomics entitled: “The Case for Mindless Economics”. In that article they made essentially two arguments. First, they suggested that neurobiological measurements, per se, lay entirely outside the province of economics. Their point was that economic theory makes predictions about behavior and thus must be agnostic about the actual machinery by which choice is accomplished. Second, they argued that while reductionist approaches which seek to link mechanistic insights to larger theoretical frameworks have been successful in the natural sciences, these same reductionistic approaches are unlikely to be able to relate natural scientific phenomena to social scientific theory. In essence, they argued that insights into biological mechanism are unlikely to have much impact on economic theory.

Several recent advances in Neuroeconomics, however, may challenge this conclusion. Glimcher and colleagues (2007), for example, measured human brain activity while subjects made choices between monetary gains of different sizes that would become available to them at different times. They found that the brain activations observed under these conditions were incompatible with an algorithmic interpretation of an important theory of intertemporal choice in use by many economists. The problem with Glimcher’s argument, however, is that the economic theory describes a mechanism that is not really hypothesized to be instantiated in the brain. Demonstrating that the mathematically specified mechanism does not exist thus only weakly contravenes the theory. A much more compelling reply to this critique would be to demonstrate that a neurobiological observation suggested a behaviorally testable modification to standard economic theory. While there is one candidate demonstration of this type emerging in the literature authored by Mauricio Delgado, Erkut Ozbay, Andrew Schotter and Elizabeth Phelps (2008), the demonstration that neurobiological data can shape economic theories of behavior remains incomplete.

The Rising Tide of Neuroeconomics

Since the publication of these and many other papers, Neuroeconomics has seen a steady growth. In 1998 less than 20 papers a year were publish that included both ‘brain’ and ‘decision-making’ as keywords. In 2008 nearly 200 articles bearing those keywords have been published. Today, a number of Centers for the study of Neuroeconomics have emerged at Universities throughout the world. Amongst the most important of these University Centers are the institutions listed in Table 1. Each of these Centers, and many others not listed here, offer both some graduate-level training in Neuroeconomics and supports independent academic researchers investigating the neural and behavioral aspects of decision-making.

In addition to these research centers, The Society for Neuroeconomics (http://www.neuroeconomics.org) now serves as a central focus for the emerging discipline. The society was founded in 2005 and hosts an annual meeting at which scholars from around the world present recent scientific findings. In 2009 the Society published, in collaboration with Academic Press, “Neuroeconomics: Decision-Making and the Brain”. This edited volume serves both as a textbook for many graduate and upper level undergraduate courses in Neuroeconomics and as a Handbook of Neuroeconomics for researchers in the field. It summarizes current advances and controversies in the field and should serve as a starting point for anyone interested in learning more about this academic discipline.

Neuromarketing

A related, although clearly distinct discipline that seems to be emerging alongside Neuroeconomics is Neuromarketing. Neuroeconomics is a purely academic discipline concerned with the basic mechanisms of decision-making. In contrast, Neuromarketing is a more applied field concerned with the application of brain scanning technology to the traditional goals and questions of interest of marketers, both those in academia and those in private industry. While these two disciplines are related, they are also very distinct. This is a distinction often overlooked by the popular media.

References

Breiter HC, Aharon I, Kahneman D, Dale A, Shizgal P (2001) Functional imaging of neural responses to expectancy and experience of monetary gains and losses. Neuron 30:619-639.

Camerer C, Loewenstein G, Prelec D (2005) Neuroeconomics: How neuroscience can inform economics. Journal of Economic Literature 43:9-64.

Delgado, MR., Ozbay, EY., Schotter, A. and Phelps, EA. (2008) Using the neural circuitry of reward to design economic auctions. Science. In Press.

Glimcher PW (2003) Decisions Uncertainty and the Brain: The Science of Neuroeconomics. Cambridge, MA. MIT Press.

Glimcher, PW., Kable, JW and Louie, K. (2007) Neuroeconomic Studies of Impulsivity: Now or Just as Soon as Possible? American Economic Review. 97(2): 142-147.

Gul, F and Psendorfer, W. (2008) The case for mindless economics. In: Caplin, A and Schotter, A (eds) The Foundations of positive and normative economics. New York. Oxford.

Kahneman D, Tversky A (1979) Prospect theory: an analysis of decision under risk. Econometrica 47:263-291.

Kosfeld M, Heinrichs M, Zak PJ, Fischbacher U, Fehr E (2005) Oxytocin increases trust in humans. Nature 435:673-676.

McCabe K, Houser D, Ryan L, Smith V, Trouard T (2001) A functional imaging study of cooperation in two-person reciprocal exchange. Proc Natl Acad Sci U S A 98:11832-11835.

Platt ML, Glimcher PW (1999) Neural correlates of decision variables in parietal cortex. Nature 400:233-238.

Shizgal P, K. C (1996) On the neural computation of utility. Current Directions in Psychological Science 5:37-43.

Internal references

Peter Redgrave (2007) Basal ganglia. Scholarpedia, 2(6):1825.

Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.

Olaf Sporns (2007) Complexity. Scholarpedia, 2(10):1623.

William D. Penny and Karl J. Friston (2007) Functional imaging. Scholarpedia, 2(5):1478.

Seiji Ogawa and Yul-Wan Sung (2007) Functional magnetic resonance imaging. Scholarpedia, 2(10):3105.

Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.

Wolfram Schultz (2007) Reward. Scholarpedia, 2(3):1652.

External links

ATR International, Computational Neuroscience Laboratories http://www.cns.atr.jp/dcn/

Baylor College of Medicine http://www.bcm.edu/ Human Neuroimaging Lab http://www.hnl.bcm.tmc.edu/overview.html

California Institute of Technology http://www.caltech.edu/ Neuroeconomics at Caltech http://www.neuro-economics.org/

University of Cape Town http://www.uct.ac.za/home/ School of Economics http://www.commerce.uct.ac.za/Economics/

Duke University http://www.duke.edu/ Center for Neuroeconomic Studies http://neuroeconomics.duke.edu/

George Mason University http://www.gmu.edu/ Center for the Study of Neuroeconomics http://www.neuroeconomics.net/

Hong Kong University of Science and Technology http://www.ust.hk/en/index.html Center for Experimental Business Research http://cebr.ust.hk/

New York University http://www.nyu.edu The Center for Neuroeconomics http://www.neuroeconomics.nyu.edu

Universiteit Maastricht http://www.unimaas.nl/ Department of Psychology http://www.psychology.unimaas.nl/

University of Muenster http://www.uni-muenster.de/en/ The Muenster School of Business Administration and Economics http://www1.wiwi.uni-muenster.de/fakultaet/

University College London http://www.ucl.ac.uk/ Gatsby Computational Neuroscience Unit http://www.gatsby.ucl.ac.uk/

University of Zurich http://www.uzh.ch/ Institute for Empirical Research in Economics http://www.iew.unizh.ch/index.en.html Research Priority Program on the Foundations of Human Social Behavior http://www.socialbehavior.uzh.ch/index.html

Neuroeconomics

Contents

Traditional Studies of Decision-Making

Some Influential Events in Neuroeconomics

Challenges to Neuroeconomics

The Rising Tide of Neuroeconomics

Neuromarketing

Recommended Books For Further Reading:

Recommended Articles for Further Reading:

References

External links

See also

Personal tools

Namespaces

Variants

Views

Actions

Search

Navigation

Focal areas

Activity

Tools