Neuroeconmics explains human behaaviour leading to decision making. Its findings contradict traditional behaviour economics but also provide substantial contribution to decision making and the way human brain works. So what's next ?
Potentially yes, I would risk saying, though not in the way that we foresee today: I don't think Neuroeconomics will radically or directly affect our current understanding and theoretical interpretation of economic systems, simply because economic behaviour is a group activity that is fundamentally uncertain. Simon has the closest-to-real interpretation of human decision as satisficing rather than optimising.
On the other side of the argument (and that is why I say I do believe Neuroeconomics will ultimately lead to change in Economic Theory) studying the behaviour of agents at the micro-level is bound to enhance our understanding of what triggers certain types of individual economic behaviour; but I think we have still a long way to go. Perhaps, by the time that impact is visible, the branch of Neuroeconomics itself will go by a different name.
Neuroeconomics may help understanding how our brain works in some sense, but I agree thatit will not replace behavioural economics, and that the 2 fields are complementary.
Neuroeconomics tells what part of the brain is involved in which kind of decision, which may help understand which decisions are related, but we still need behavioural economics and decision theory to understand how people process information and make decisions. Dan McFadden (economic Nobel prize, 2000) is particularly interested in this topic. You may have a look on the paper " Process and context in choice models" published in Marketing letters 2 years ago. No need to rely on neuroeconomics to analyze the decision process.
Although many researchers and theoreticians are quite interested in this new proposal and its potential, there's still a long way to go. From the philosophy of science you may have two possible analysis: Kuhn's paradigm point of view (1970), and the latter Lakatos' proposal on scientific research programmes (1978). To follow Kuhn's point of view, it would be necessary for neuroeconomics to provide new tools of research, that answer questions the «existing» paradigm is not able to respond, leading more and more research into new methods to solve questions of interest. However, there might be a limitation nowadays since neuroconomics investigation is very expensive, limiting the possibility of larger samples. Nevertheless, nothing is already told or concluded.
On the other hand, by Lakatos' point of view, there's still a couple of possibilities for this new research: it might be just represent a «local divergence», or even generate a new field of research outside the hardcore of neoclassical programs. Local divergence implies that it is still possible for neoclassical economics to develop an extension of its ad-hoc hypothesis, inserting the new tools to the previous research. New research programs implies the appearing of a completely new field, that may compete with the neoclassical research program.
It might be too soon to attempt any kind of prediction (in my point of view) of the possible future in economic theory. In addition, there are even more kind of new research proposals (like theory of chaos, experimental economics, evolutionary game theory, disequilibrium economics, and inside the neoclassical research itself, etc.) that might represent new frontiers and possibilities to lead the future economics.
References.
Kuhn, T. S. (1970) The structure of scientific revolutions. The University of Chicago Press.
Lakatos, I. (1978) The methodology of scientific research programmes. Cambridge University Press.
This is very interesting question. The recent development in behavioural aspects has been emerged in response to the debate about problem with existing established behavioural models.
In response to the first answer that people are not computer in relation to computational approach to neuro-economics, I think most concepts in computer science is originated from psychology, such as enforcement learning technique in machine learning. Techniques like Agent Based Modelling are based on more realistic assumptions about human behaviour. For example, methodology in this area is based on heterogeneous and bounded rational agents.
These areas together with neuro-economics are new compare to other established methods. However, it would equally be naïve to reject them completely.
I agree with you Mona. Of course you can't reject any kind of theory, especially when there are many published papers and seious scientific research. I strongly believe that neuroeconomics has a substantial contribution to make and even if it doesn't entirely alter economic theory, it will definitely change the way of understanding human behavior. It is the future.......
I just returned from the 2014 Society for Neuroeconomics (SfNE) conference in Miami, where I presented a poster, "Neuroeconomics and revealed-preference theory as synergistic cornerstones in economics: Linking neural and choice data may enable a novel self-regulatory policy for preventing asset-price bubbles", which was co-authored by Daniel J. Acland, who is an economist on the faculty at the Goldman School of Public Policy, UC Berkeley. I will present an updated version of this poster at the 2014 Society for Neuroscience (SfN) conference in Washington, DC on the afternoon of November 17 (Presentation Number: 459.03; Posterboard Number: SS43). All who would like to discuss the asked question or my poster are welcome to stop by during this poster session. I believe that our understanding of some specific topics (e.g., asset-price bubbles) will be improved by a synergism between neuroeconomics and traditional economic theory in an approach that could be called neuroeconometrics. Research in this area may yield a neuroimaging-based approach to financial-system regulation. Potential research directions will be discussed in the attached proposed seminar that Dan Acland and I are organizing at UC Berkeley. If it receives administrative approval, this seminar, titled "Neuroeconomics Interventions to Reduce Asset-Price Bubbles Associated with Animal Spirits", will be offered in the Spring 2015 or Fall 2015 semester, when it will involve 3-5 meetings. Anyone (e.g., faculty members, postdocs, or students) interested in attending this seminar may email me ([email protected]), so that I could add you to the list of participants. An affiliation with UC Berkeley is not required for attending the seminar. My posters, titled "Neuroeconomics of asset-price bubbles: A potential role for herding", that I presented at the 2013 SfNE and SfN conferences are available at my crowdfunding website - http://www.gofundme.com/9zcxe4 - which was set up to seek funding for the above research program. This website also will make available my 2014 SfNE and SfN posters after the SfN conference in November.
Many of the reinforced learning models remind me of John Elster's comment that "men are not happiness machines". Given a fixed reward system I am sure they work well enough. But they do not tell us what is valued and hence what rewards will motivate. In general I prefer some economic theorisation to direct biological/algorithmic analogies but this may reflect my own skill set more than problem at hand.
Yes, William Forbes, your last sentence about an interaction between economic theorisation and "biological/algorithmic analogies" represents part of the synergism that I proposed between traditional economic theory and neuroeconomics, possibly leading to the development of neuroeconometrics as described in my attached seminar proposal and Society for Neuroeconomics conference abstract. Anyone who would like to participate in the seminar may email me at [email protected].
Will Neuroeconomics change traditional economic theory?
This question is a wolf in sheep’s clothing.
It is certainly the case practically that some areas of economics are not going to be largely modified by knowledge of the biological mechanisms of individual human decision making. However, there certainly are areas where neuroeconomics should be able to significantly improve predictions of individual decision making in the short-term.
At a theoretical extreme, assume a full computational model of the biological mechanisms of human decision making. Definitionally, this model would provide the most accurate prediction of behavior possible.
Given this incredible model, would it be utilized? No. The computational complexity of this model would make it non-viable for any concrete purpose. Simply, the costs of using such a model would certainly restrict its practical use.
Practically, we have to find a middle-ground. This is where neuroeconomics is likely to be able to provide concrete contributions to improving models of human behavior. A few theoretical examples are:
1) As we identify the actual mechanisms of human decision making, we reduce the dimensional space of viable economic models. As an example, let us assume that we have two competing models of a specific choice behavior. Mathematically it is possible for two disparate models to arrive at comparable predictive power. Knowledge of the biological information processes of that decision making should allow us to distinguish between these two models. Such specification of viable models will facilitate advancing those models.
2) In addition, knowledge of the biological processes of decision making will allow us to identify and access significant ‘hidden’ variables that are unavailable from behavior alone. For example, a wealth of studies suggest that the ventromedial prefrontal cortex is involved in processing the subjective value (‘utility’) of a presented stimulus, and have even shown that knowledge of the relative activation of this region across presented stimuli can predict choices between those stimuli.
As such, this would suggest that we have access to the idiosyncratic valuations of an individual. Such access could conceivably lead to significant improvement of our understanding of the processes through which individuals determine their subjective valuations. In other words, rather than having to rely solely on choice behavior to infer the way in which an individual interprets the value of a stimulus, we can access that information directly.
The incorporation of these variables into economic models should lead to significant improvements in their predictive power for individual behavior.
With a bit of a jump, I believe that these improved models of individual behavior (micro) will then lead to improvements of the behavior of groups of individuals (macro).
Neuroeconomics is a young field, and the brain is incredibly complex. Give us ten more years.
A quick note: simply showing that a brain area is ‘active’ during a particular behavior is the least specific form of neuroimaging analysis. One of the aims of neuroeconomics is to identify the neural mechanisms behind decisions - the specific information processes and transformations responsible for our choices. The localization of activity across tasks facilitates early investigations of this form, but it is far from the endpoint. There are neuroimaging techniques that allow for much greater specification of the information processes, and additional techniques that allow us to fully-determine those processes (for a clear example, recordings of individual or groups of neurons). We can even show the causal role of the information through a host of techniques (ranging from TMS to micro stimulation). There are no clear hurdles, other than time and complexity, to our eventually producing a full computational model of the biological mechanisms of human decision making.
This is an interesting discussion for me, as I considered the possibilities of decision neuroscience almost a decade ago extensively, from a systems-theoretical point of view. Like earlier process-tracing studies in behavioural sciences, neuroscience might definitely boost our understanding of how the human brain works during a choice process. Particularly, it has already identified locations within the brain where mental activity takes place and will definitely refine and extend these. Presumably, in the future it will also reveal the temporal sequence in which this occurs. Suppose that the relevant processes like information storage and retrieval, appraisal, evaluation etcetera occur in the same dedicated areas of the brains of all humans, and that neuroscience would be able to identify these locations unambiguously. Then, we would have a much better understanding about the thought process, i.e. how the brain processes transform the supplied information (input) into a choice decision (output). Would this yield a generic choice model that is superior to the current models to predict economic behaviour? In my opinion this is not the case because it does not supply knowledge about the function and purpose of the choice process, i.e. what transformations (algorithms etc.) are applied successively to the choice-option-information input to arrive at the choice-decision output and why does the brain so. See for an extensive elaboration of this line of thought Van de Kaa, 2008 (Extended Prospect Theory…, Chapter 2, TRAIL RS, Delf) as summarized in Van de Kaa, 2010 (Prospect Theory and Choice Behaviour Strategies…, EJTIR), which are both accessible on Research Gate. This implies that whatever exciting insights neuroscience will yield in the foreseeable future, applied scientists (like economists and transportation researchers) who want to improve their models are doomed to keep on hypothesizing about functional-descriptive human choice models and testing whether these offer acceptable predictions. In my opinion the seminal article of Friedman (1953, The methodology of positive economics. In: M. Friedman (ed), Essays in Positive Economics, pp. 3-43, Chicago, University of Chicago Press) still offers useful insights into the possibilities and limitations of such humble applied research efforts.
This is a legitimate question that deserves answering without a need for speculating about what the question might be other than what it actually is. I will give a specific example that supports my 10/2/14 answer's belief that a neuroscientific approach, which could be called neuroeconometrics, may improve our understanding of specific topics in economics (e.g., asset-price bubbles). Traditional economic modeling inadequately accounted for the destabilizing risk of contagion in market economies (Stiglitz JE. The lessons of the North Atlantic crisis for economic theory and policy. In: What Have We Learned? Macroeconomic Policy after the Crisis. Akerlof G et al., eds. Cambridge, MA: MIT Press, 2014, pp. 335-347). Functional magnetic resonance imaging (fMRI) studies have yielded an emerging pattern of results which suggests that neurocircuitry involving evolutionarily ancient areas may have a key role in mediating contagion (i.e., "herding" or decision making that follows others' decisions). Herding-related fMRI activations were found in nucleus accumbens (i.e., archistriatum; Klucharev et al., 2009; Burke et al., 2010; Zaki et al., 2011), hippocampus (i.e., archicortex; Edelson et al., 2011), and amygdala (Burke et al., 2010; Edelson et al., 2011). On the other hand, many deliberative processes, such as calculating and deductive reasoning, are more related to lateral neocortical activations, particularly in the frontoparietal network (Corricelli and Nagel, 2009; Shirer et al., 2012). This pattern of results supports a neuroeconomics-based hypothesis of asset-price bubbles. In this view, evolutionarily ancient or new neurocircuitry drives, respectively, herding-related decision making during bubbles or deliberative decisions in non-bubble periods of financial-market activity. Consistent with this hypothesis, Smith et al. (PNAS 111:10503-10508, 2014) found a "neural metric for irrational exuberance" (p. 10506) consisting of nucleus accumbens activity that, when calculated as a moving average across all lab asset-market subjects, tracked bubble-related price changes and predicted crashes. This finding is an initial step toward a proof of concept for the presently proposed neuroimaging-based approach to financial-system regulation. The above neuroeconomics-based hypothesis of bubbles yields the testable prediction that, relative to non-bubble periods of market activity, participants in real financial markets should show low decision-related lateral neocortical activity during bubbles. Field studies could test this hypothesis with functional near-infrared spectroscopy (fNIRS), which is a portable, relatively inexpensive neuroimaging method that is largely limited to monitoring activity in neocortical areas. Suitable subjects for these field studies would be individual investors who are prone to mistaken investment decisions that may play a substantial role in the driving of asset-price bubbles (Griffin et al., J. Finance 66:1251-1290, 2011). This research may yield a bubble-related biomarker that is detectable by fNIRS in real time. For example, the above neuroeconomics hypothesis of bubbles would be supported by finding low trade-related lateral neocortical activity during asset-price bubbles that later crash. This low lateral neocortical activity could thereafter function as a biomarker for alerting government regulators to implement countercyclical measures (e.g., adjusting caps on loan-to-value ratios for mortgages and increasing capital requirements for banks) in order to prevent major bubbles and crashes. The investment of time and resources in this research program is justified due to the potentially large payoff of preventing post-crash prolonged recessions. A more detailed description of the proposed research program is in the attached updated proposal for a neuroeconomics seminar at UC Berkeley in 2015. Scientists or scientists-in-training who would like to participate in the proposed seminar are welcome to email me at [email protected]. A complete list of the references cited above is available at: http://www.gofundme.com/9zcxe4. Happy New Year to All.
http://www.gofundme.com/9zcxe4
I don't think so, just as the birth of the computer age gave rise to simulations as a third way of "doing science"; besides one rooted in induction (phenomenological) and one rooted in deduction (empiricism). So in an analogous way, neuroeconomics will give rise to tools to unravel the black box that is at the core of homo economicus. Behavioral economics has unraveled psychological aspects that explain irrationality, neuroeconomics will unravel the neural-cognitive pathways linking economic decisions to economic behavior. The greatest impact on economics would be in the resurgence of interest in cardinal utility. Giving rise to clearer delineations between conditions and cases which up till now remain fuzzy category-wise.
There's much that's useful in the answers and comments already offered. Here is a thought (implicit in some of the answers): Your reference to 'traditional economic theory' is potentially ambiguous between (at least) economic theory understood as normative, and as predictive. For normative theory, aimed at determining the optimal choices or allocations under various conditions, it's plausible to argue that no empirical finding, including findings about the brain and how choices by real agents are actually made, has any bearing at all. In the predictive case, however, where we're after a theory of what agents (including people) actually do, it's hard to see that finding out more about how they actually work won't make a difference. In that case the relevance of neuroeconomics is as a continuation and extension of the research programme of behavioural economics. (That is, it's behavioural economics that looks 'under the hood' to see what is going on. And it is much easier to see how neuroeconomics can influence behavioural economics, because results about neural encoding and computation of value can help decide between incompatible behavioural theories that are on a par in their capacity to explain choice data.)
Paul: Neuroeconomic really only deals with decision-making. It is helpful when studying consumer decision-making or the decisions made by suppliers.
Traditional economic theory is much broader and deals with macro and micro economics, as well as fiscal and monetary policy.
I think neuroeconomics is a very small subset of Economics.
- Badges
- Science topic