PsyFi Search

Tuesday 9 February 2010

The Neuroeconomics Revolution

A Marriage Made in Heaven – or Hell?

A recent development in economics sees the combining of neurology, psychology and economics in an attempt to reduce economic behaviour to brain function and to predict market behaviour from observable brain patterns. Its aim is to glue together a subject that can’t predict human behaviour from analysis of the brain with a subject that can’t predict human behaviour from analysis of people to a subject that can’t predict human behaviour from analysis of economic data.

Welcome to the Neuroeconomics Revolution.

The Busted Flushes of Psychology and Economics

Psychology spent a large part of the last century stuck in a behaviourist dead-end, carrying out endless experiments on animals in an attempt to explain all human behaviour in terms of externally observable responses to equally observable stimuli: think ringing bells causing salivating dogs or rising markets causing manic investors. The net result of this was that the subject ended up befuddled by mice running the wrong way around mazes and found itself generally regarded as the extreme paramilitary wing of the pigeon fanciers association.

Meanwhile economics, the study of how human financial systems operate, also proceeded on the basis that how humans actually behave was irrelevant and arrived at a set of explanations that defied both logic and the evidence of real markets. Yet even as economics has reluctantly faced up to the need to involve psychology in its models so psychology is beginning to recognise that understanding people requires a more detailed look at the way the brain actually works. Taken together we’re witnessing the creation of a new subject.

Neuroeconomics

Neuroeconomics is nothing more or less than the attempt to relate the now observable functioning of the brain, as provided by neuroscientific techniques, with the various models of economics. Advances in brain scanning techniques permit researchers to subject innocent participants to endless pointless questions while inspecting how their brains grapple with the problems. There is no escape.

As you might imagine, the marketeers of the world have leapt on this idea. Martin Lindstrom relates in Buyology how the inspection of people’s brains reveals what more observant marketing experts and psychologists have suspected for quite some time: asking people what their preferences are is largely a waste of time and dollars. The reason for this is quite simple – they don’t actually know, because the bits of their brains that decide whether they like something or not aren’t available for conscious inspection.

To give a single example from Lindstrom’s book, the analysis of brain processing gives the lie to the famous Pepsi taste test. In Neural Correlates of Behavioral Preferences for Culturally Familiar Drinks the researchers showed that in a blind tasting session more people preferred Pepsi over Coke despite the reverse being true when the test was unblinded. What the study of neurology showed is that the brain instinctively and initially prefers the sweeter taste of Pepsi but that emotions then kick in to overrule the initial preference in favour of Coke. Presumably this is to do with the effects of branding and other cultural preferences built over a lifetime.

Reductio Ad Absurdum

While it may seem instinctively obvious that wider economic behaviour can be reduced to individual personal behaviour and interactions and that these can be reduced to the way the brain works this is a fairly controversial idea. To start with many psychologists are opposed to such an application of reductionism, the idea that complex human behaviour can be explained by the firing of neurons and, to some extent, they’re clearly right.

For example, we know the brain is an adaptive learning machine – examples of people who’ve been born with bits of their brains missing yet have led virtually normal lives show exactly how plastic the function of the early, developing brain is. What we become is heavily dependent on what we experience. So it’s understandable that many social scientists don’t think that neurology can explain how we become what we are.

Despite this, it’s hard to argue that the developed brain doesn’t represent the way we actually are. If that’s true then neuroscience ought to be able to help us understand some of the behavioural issues that perplex psychologists. As these are often the same issues that puzzle economists from a behavioural finance perspective there seems to be a fairly simple line to be drawn from psychology to neurology.

Economic Implementation

However, many economists don’t accept that economic modelling can gain anything from neuroscience. Their argument, to simplify vastly, is that economic models are not dependent on how they’re implemented – what matters is that the models are shown to be true. Philosophers also argue in this way – so, for instance, there seems to be no restriction on consciousness being exclusively human and if it can be “implemented” on multiple different platforms – in the human brain or on complex supercomputers or in those strange energy beings dedicated to killing hapless extras in Star Trek – then there need be no direct link between the brain and advanced behaviour. You know, stuff like buying the wrong stocks at the wrong time for the wrong reason.

With which there is just the tiniest problem: this is all theoretical and until someone actually finds consciousness in something other than the human brain or its co-evolvees, it's simply speculation. If economic systems are dependent on human behaviour and human behaviour is dependent on the way the human brain is built – which all seems rather likely – then it’s perfectly possible to make inferences from neurology to economy via psychology.

Plastic Economic Behaviour

In addition even the most recalcitrant of psychologists will concede that people’s behaviour changes depending on what they experience. So someone who endured the cataclysmic collapse of Wall Street in 1929 and suffered the grinding poverty and desperation of the Great Depression is likely to behave differently than someone brought up in the go-go Fifties or the gung-ho Nineties. Take any particular cohort of people from a specific background and subject them to the same experience then they’ll likely behave differently from other such groups.

So how likely is it that the economic models that don’t take human behaviour into account will generate realistic results? Well, of course, those models that do take human behaviour into account probably won’t do much better, but that’s because psychologists don’t yet know how to reduce human behaviour to typical neurological patterns and mechanisms.

Weirdness Personified

However, they’re working on it. By comparing brain activation on certain tasks of people with and without specific types of brain damage it’s painstakingly possible to build up a picture of what bits of the brain are engaged in what bits of processing. Researchers would prefer to stick electrodes straight into people’s brains to directly see the processing going on but this is generally frowned upon by boring ethics committees who are comprised of figures who probably reckon that they’ll be amongst the first to be plugged in and zoned out if it’s ever permitted.

So, failing this option, psychologists stick electrodes into the brains of various small animals, thus getting their revenge on those deceitful maze running mice, or use different sorts of scanners that give indirect access to human neural processing. As a result of this work we can sort of figure out which areas are involved in the various sorts of experiments that economic researchers find interesting. However, the difficulty is in interpreting these in terms of “higher behaviour” and – perhaps an even worse problem – making sure that the experiment is socially relevant. Too many psychology experiments put people into situations that are just plain weird in an attempt to control the variables: the results are invariably exciting and fascinating – and ultimately meaningless.

Monkey Business

Although all of this is still in its early days there are some interesting suggestions emerging about how various types of behaviours emerge from brain function and how these do – or don’t – map to various economic theories. Paul Glimcher, for instance, has shown that monkey neurons can appear to achieve a Nash equilibrium, basically the economically optimal algorithm for the task at hand: Glimcher's also written an excellent overview of neuroeconomics. Meanwhile trust games are a particular favourite of investigators because these allow hypotheses to be generated based around, say, utility theory which can then be tested.

As we’ve seen before, in Stocks Aren't Snakes, this work is suggesting that much of our seemingly irrational behaviour comes out of the new brain areas developed to support advanced human social behaviour. So while the older areas are straightforwardly selfish maximisers the newer areas will cause us to turn down offers that are in our interests because of concepts of social justice. All of which suggests that lizards might be better investors than humans. This isn't uncontroversial, mind you: a lot of researchers think it's poppycock.

Neuroeconomics is only likely to grow in interest to economists and marketeers alike as the technology to scan the brain and interpret its results increases: it seems to offer new ways of modelling old problems to come up with existing results. I’ll revisit this shortly to look at what, of better or worse, this may actually mean.


Related Articles: When A Dollar's Not Just A Dollar, Stocks Aren't Snakes, B.F. Skinner's Stockmarket Slot Machines

3 comments:

  1. I don't think it is possible to know enough about the brain for this sort of thing to pay off for investors.

    I also don't think it is necessary.

    My view is that valuations tell you what you need to know about the non-rational side of investing.

    We are trying to solve problems that do not exist as a means of avoiding acceptance of the solutions to the ones that do exist that are already available to us.

    Rob

    ReplyDelete
  2. Thanks for the good writing.
    I just went to the CNS site and read a bit.
    Rob, do not preclude the possibility that marketing any kind of positive test result will happen. Monies will be allocated in budgets to get a perceived edge, necessary or not.
    I don't think it has the same implications gene therapy/cloning do, at least not yet.
    Just what is Dr. Andrew Lo doing these days, in addition to AlphaSimplex, and who are the leaders in this field?

    ReplyDelete
  3. "Neuroeconomics is only likely to grow in interest to economists and marketeers alike as the technology to scan the brain and interpret its results increases: it seems to offer new ways of modelling old problems to come up with existing results"

    While this statement maybe quite true, it might also be worthwhile to consider the effects of neurofeedback and regulation methods on human behavior, under circumstances which causes economic havoc, for example, gambling, and whether these can be identified and regulated neurologically as well. Applications like these might make the field worth pursuing, especially when we require an "objective" feedback of what is going on, rather than the subject himself expressing or evaluating his behavior in a given circumstance.

    ReplyDelete