A neuroeconomic framework

In a recent review article in Nature Reviews Neuroscience, Antonio Rangel, Colin Camerer and Read Montague suggest a framework for neuroeconomic research. Indeed, the very core of its idea is simple, but not simplistic. After reading the article, I think it will provide a useful reference for future research into neuroeconomics, aka value-based decision making. I’ve made a copy of the model here for you to see:

The caption reads:

Basic computations involved in making a choice. Value-based decision making can be broken down into five basic processes: first, the construction of a representation of the decision problem, which entails identifying internal and external states as well as potential courses of action; second, the valuation of the different actions under consideration; third, the selection of one of the actions on the basis of their valuations; fourth, after implementing the decision the brain needs to measure the desirability of the outcomes that follow; and finally, the outcome evaluation is used to update the other processes to improve the quality of future decisions.

In my own emerging work on this arena, I am trying to combine this with recent advances cognitive neuroscience. First, the advances in imaging genetics, i.e., the knowledge and study of how genetic variance leads to specific changes in neurotransmission, which in turn may affect cognition, emotion and behaviour. Second, the advances in the cognitive neuroscience of ageing, i.e, the relationship between age-related changes in brain structures and functions, and mental alterations.

Briefly put, in a just submitted manuscript, I suggest that the Rangel-Camerer-Montague framework can serve as a model for looking at genotype and age effects. This leads us to three advances: first, it provides a better way to illustrate and understand the minute details of the preference and decision making systems. Second, it serves as a demonstration that individual (and intra-individual) differences must be taken into account. The “economic agent” is not a homogenous subject, but an agent that differs from person to person and with persons over time. Finally, it may also serve as a framework for identifying potential ways to induce alterations in the systems, e.g., through medical intervention. More on this story later, given that the manuscript is accepted 😉 For now, here’s an illustration of how genotype (exemplified through COMT, MAO-A and 5-HT) and age effects may expand the model. Of course, this is only scratching the surface, but I hope you’ll see what I mean.

This is an extended version of the Rangel-Camerer-Montague model. Within each processing node, two dimensions are added, here exemplified with the three primary nodes. The genotype dimension is a categorical variable that divides subjects into two or three classes, while the age dimension is continuous (inset, top left).

-Thomas

Read Full Post »

Serotonin and value-based decision making

Can antidepressive medicine alter your decision behaviour? A recent paper in Science now demonstrates that alterations in subjects’ serotonin levels leads to significant changes in their decision making behaviour. In the study, subjects were set to play the Ultimatum Game repeatedly. Subjects had to do the task two times at two different days, and at one of the days they were administered an acute tryptophan depletion (ATD), i.e., their serotonin levels would drop for a period of time. The design was double-blind and placebo controlled.

The Ultimatum Game is an experimental economics game in which two players interact to decide how to divide a sum of money that is given to them. The first player proposes how to divide the sum between themselves, and the second player can either accept or reject this proposal. If the second player rejects, neither player receives anything. If the second player accepts, the money is split according to the proposal. The game is played only once, and anonymously, so that reciprocation is not an issue.

What the researchers found was that the ATD led subjects to reject more offers, but only unfair offers. That is, ATD did not interact with offer size per se, and there was no change in mood, fairness judgement, basic reward processing or response inhibition. So serotonin seemed to affect aversive reactions to unfair offers.

The study is a nice illustration of how we now are learning to induce alterations in preferences and decision making. Along with other studies using, e.g., oxytocin to increase trust in economic games (see also my previous post about this experiment), one may expect that increasing the serotonin level may actually make subjects less responsive to unfair offers.

This knowledge is also important to learn more about, as it poses a wide range of ethical problems. If our preferences and decisions are really influenced by these stimuli, can this be abused? It should be mentioned that many of these substances are not necessarily detected (oxytocin is odourless), so we may be influenced without our consent or knowledge. The wide applicances could include casinos, stores (e.g. for expensive cars), dating agencies and so on. If we did not accept subliminal messages in ads, how can we accept this?

-Thomas

Read Full Post »

Oxytocin: the direct route to altruism?

In relation to our previous and well-visited post about oxytocin, we should mention a new study that uses this very substance in a neuroeconomic set-up. In the study, recently published by Neuron, and headed by Baumgartner et al., it was found that the administration of oxytocin affected subjects’ in a trust game. In particular, it was found that subjects that received oxytocin were not affected by information about co-players that cheated. Or, as put in the abstract:

(…) subjects in the oxytocin group show no change in their trusting behavior after they learned that their trust had been breached several times while subjects receiving placebo decrease their trust.

That is extremely interesting. This suggests that oxytocin, a mammalian hormone + neurotransmitter that is known to be related to maternal behaviour and bonding, also is modulating social trust. So the brain link is obvious. But what happens in the brain when oxytocin is administered during the trust game?

This difference in trust adaptation is associated with a specific reduction in activation in the amygdala, the midbrain regions, and the dorsal striatum in subjects receiving oxytocin, suggesting that neural systems mediating fear processing (amygdala and midbrain regions) and behavioral adaptations to feedback information (dorsal striatum) modulate oxytocin’s effect on trust.

So oxytocin reduces fear and aversion responses, and this leads to the lack of effect in responding to cheaters. Excellent, why not use this for treating anxiety, phobia and other fear-related problems? Sounds promising, and yet other more ethically problematic issues remain to be resolved. Think, for example, about whether oxytocin makes us more susceptible to gambling, shopping and marketing effects? Or what if it may work as the first scientifically proven aphrodisiac? What if your next pick-up line would be “Hi, I’m Thomas, how are you” just followed by a few ‘puff-puff’ sounds.

Joke aside, studies like this demonstrates that emotions and decisions are often influenced by factors not consciously available, or at least only partially so. As the marketing industry is increasingly interested in multi-sensory inventions, oxytocin may be the next step in this endaveour.

-Thomas

Read Full Post »

Oxytocin is the window to the soul

Research on the role of oxytocin, a neuropeptide, in social cognition has generated much interest during the last few years. We have earlier written about oxytocin’s role in social attachment; together with vasopressin, another neuropeptide, oxytocin is thought to be critical for linking social signals to structures in the mesolimbic part of the brain responsible for forming social attachment and pair bonding. In a study published in Nature Ernst Fehr and his group demonstrated that injecting people with a spray of oxytocin increases trust.

Now, in a pretty remarkable new study published in Biological Psychiatry, German researchers show that injecting subjects with a whiff of oxytocin will also improve be ability to infer, based just on eye cues, what a person is thinking about. Here’s the abstract:

Background

The ability to “read the mind” of other individuals, that is, to infer their mental state by interpreting subtle social cues, is indispensable in human social interaction. The neuropeptide oxytocin plays a central role in social approach behavior in nonhuman mammals.

Methods

In a double-blind, placebo-controlled, within-subject design, 30 healthy male volunteers were tested for their ability to infer the affective mental state of others using the Reading the Mind in the Eyes Test (RMET) after intranasal administration of 24 IU oxytocin.

Results

Oxytocin improved performance on the RMET compared with placebo. This effect was pronounced for difficult compared with easy items.

Conclusions

Our data suggest that oxytocin improves the ability to infer the mental state of others from social cues of the eye region. Oxytocin might play a role in the pathogenesis of autism spectrum disorder, which is characterized by severe social impairment.

Clearly, this suggests that oxytocin not only modulates mesolimbic brain structures. Earlier studies have implicated the fusiform face area and superior temporal sulcus in extracting social information from facial perception. May oxytocin also impact on these structures? Whatever turns out to be the case, I imagine that no politician or CEO will ever sit down at the negotiation table again without their trusty bottle of oxytocin.

-Martin

Read Full Post »

Next Im-Gen meeting

The next International Imaging Genetics Conference is opening its doors now for registration. The third year in a row, building on two successful conferences, this third meeting will also house two separate workshops: one on brain imaging for geneticists; and one on genetics for brain imagers. All in the spirit of crossing the bridge between genetics, brain imaging and statistics. As this course was brilliant last year, I’m hoping to attend in January 2007, too.

Here is the announcement:

The First and Second International Imaging Genetics Conferences were held to bring together national and international experts in neuroimaging, genetics, data-mining, visualization and statistics. Targeting physicians and scientific researchers, this annual conference features presentations from investigators world-wide and held in-depth discussions within the emerging field of Imaging Genetics. Given the known importance of both genetics and environment in brain function, and the role of neuroimaging in revealing brain dysfunction, the synergism of integrating genetics with brain imaging will fundamentally change our understanding of human brain function in disease. To fully realize the promise of this synergy, we must develop novel analytic, statistical, and visualization techniques for this new field.

This international symposium was held to initially assess the state of the art in the various established fields of genetics and imaging, and to facilitate the transdisciplinary fusion needed to optimize the development of the emerging field of Imaging Genetics. The Third Annual International Imaging Genetics Conference will be held on January 15^th and 16^th, 2007 at the Beckman Center of the National Academy of Sciences in Irvine, CA. We look forward to seeing you at this exciting upcoming event.

Monday January 15th:

• Nicholas Schork, UCSD “Multivariate Analysis of Combined Imaging and Genomic Data”
• Eleazer Eskin, UCSD “Analysis of Complex Traits Through Intermediate Phenotypes.”
• Tom Nichols, University of Michigan “Statistical Challenges & Opportunities in Imaging Genetics”
• Fabio Macciardi, University of Toronto “Integrating Imaging Genetics Methods in Schizophrenia.”
• David Goldman, NIAAA “Genes and Neurobiologies in the Addictions”
• David Goldstein, Duke Institute for Genome Sciences and Policy “Neuropsychiatric pharmacogenetics”
• Daniel Weinberger, NIMH/NIH: TBA

Tuesday January 16th:

• Joseph Callicott, NIMH “Does risk for schizophrenia arise from multiple genes in vulnerable pathways? Evidence from DISC1 and FEZ1”
• Lisa Eyler, UCSD “Genetics of Brain and Cognition: A Twin Study of Aging”
• Fei Wang, Peking University “Neuregulin 1 Genetic Variation and anterior cingulum integrity in schizophrenia and in health.”
• Andreas Meyer-Lindenberg, NIMH/NIH “Genetic characterization of prefrontal-subcortical interactions in humans.”

**New for 2007** Sunday January 14th:

*** Half-day Workshop tutorials will be offered the day before the conference at the Beckman Center- see website for details***

Workshop 1: What Geneticists need to know about Brain Imaging
Workshop 2: What Brain Imagers need to know about Genetics

Registration and conference information can be found at the conference website

-Thomas

Read Full Post »

The making (and tailoring) of instincts

What is the nature of instincts and inborn behaviour? The cover article of this week’s issue of Current Biology is an article by Kim et al. on how the central nervous system produces inborn behaviour. The researchers found that the innate behavior is initiated by a “command” hormone that orchestrates activities in discrete groups of peptide neurons in the brain (peptide neurons are brain cells that release small proteins to communicate with other brain cells and the body). The action of the hormonal influence, however, was not an all-at-once phenomenon. The ecdysis-triggering hormone or ETH, activated discrete groups of brain peptide neurons in a stepwise manner, making the fruit fly perform a well-defined sequence of behaviors.

Says Michael Adams, the research team leader:

Our results apply not only to insects; they also may provide insights into how, in general, the mammalian brain programs behavior, and how it and the body schedule events. By understanding how innate behavior is wired in the brain, it becomes possible to manipulate behavior — change its order, delay it or even eliminate it altogether — all of which opens up ethical questions as to whether scientists should, or would want to, engineer behavior in this way in the future”

An instinct is an inborn disposition for a specific kind of behaviour. Although we often regard instinctual behaviours to be in the domain of non-human animals, humans also have well-known instincts, the most known are found in infants. However, many researchers suggest that other behaviours — such as altruism, disgust, face perception, and language acquisitions are also instinctual, at least in part so.

Would the present result mean that behaviour — let’s start with instinctual behaviours — could be edited? Adams seems to think so, and in the Drosophila it certainly seems to be the case. Since the steps leading to the well-orchestrated instinctual behaviour were identified, and the neurochemical properties of these circuits are — or are soon — known, it should be a trivial thing to alter the behaviour at any stage during the process. What about human behaviour, then? In principle, the same idea should apply.

Complaints will certainly arise that “human behaviour is so much more complex than instincts”. Indeed it is. But recall studies of voluntary action, of subliminal perception and of automaticity (e.g. riding a bike). These behaviours resemble instinctual behaviour a lot. They are not under conscious control, or are at least not initiated by such. In this sense, it might be that even human behaviour can be edited in much the same way as the Drosophila.

Admitted, this study only focuses on the ecdysis sequence behaviour in an insect. So we need more advanced (still instinctual) behaviours to believe the story. Replications & variations as always… But hey, maybe this will end up as the next cure for odontophobia…?

-Thomas

Read Full Post »

MAOA and the risk for impulsivity and violence

Violence and criminal behaviour is today thought to involve a series of complex interactions between heritable and environmental factors. Centuries of debate of the relative contribution of nature and nurture have not reached anything resembling a solution, and even today we can find ardent proponents and defenders of each extreme view (see Steven Pinker on this, PDF).

While violence and crime has been part of all recorded history, the society’s understanding of the underlying causes of these acts and how they should be dealt with have changed over time. In modern times, we also see a wide variety of legal practices in dealing with criminals and violators: from the death penalties and multiple life sentences in the US, Russia and other countries, to briefer treatment sentences in Europe. These different societal solutions build – explicitly or implicitly – upon what causes violence and criminal acts, and how they should, if at all possible, be treated.

It would be no understatement to claim that the biological explanation of violence and crime has not been fully implemented (nor understood) by law makers or enforcers. Just as you could say about the society in general: aside from specific demonstrations of how violent offenders have larger or smaller neural damage, little is known about the biological properties of violence. Not that the literature has been flourishing with articles demonstrating such relationships. It hasn’t. Until now, where recent studies report detailed analyses of how genes and environments alter the brain’s workings to make people more or less prone to violence, impulsive acts and criminal behaviour.

In a most interesting paper (PDF) published in PNAS, a team of researchers from Austria, Italy and USA headed by Andreas Meyer-Lindenberg have uncovered neurobiological factors that contribute significantly to violence in humans. The team studied the normal allelic variation in the X-linked monoamine oxidase A (MAOA) gene, a gene that has also been shown to be associated with impulsive aggression in humans and animals.

In the study both structural and functional MRI methods were applied. First, the researchers asked whether the low expression variant of MAOA, known to be associated with increased risk of violent behaviour, would predict differences in the size of limbic structures such as the amygdala. Indeed, what they found was that the low expression MAOA predicted limic reductions, as can be shown from the figure article

Structural reductions in limbic and paralimbic regions due to genotype. The size of both the amygdala and cingulate cortex are predicted by benotype. The low expression MAOA have significantly reduced volumes of these structures, compared to the high expression MAOA group.

Second, the team studied how these structures worked using two fMRI paradigms. The first task was a facial expression matching task, a task known to involve the amygdalae. The amygdala activation was significantly influenced by genotype: the low MAOA group displayed higher amygdala activation and at the same time lower activation in cingulate cortex subregions, as well as left orbitofrontal cortex and left insular cortex – all brain regions implied in emotion processing.

Regions involved in facial expression matching (click image for larger version). As you can see from the graphs, there is a genotype-by-gender interaction.

The second task was an emotion memory task, where subjects were asked to encode and recall aversive (compared to neutral) valenced information. Here, the results pointed to a significant genotype-by-gender interaction effect, in that men with a low MAOA version showed increased reactivity of the left amygdala and hippocampus during recall. No such relationship was found for women.

Interestingly, the researchers also found a tight relationship between gender and genotype during the first volumetric study. Here, low-MAOA males showed increased orbitofrontal volume bilaterally, while no such relationship was found in females. In this sense, the MAOA allelic variances seem to influence males most.

Finally, Meyer-Lindenberg and his co-workers draw the lines to other studies relating MAOA variance to a highened sensitivity in low-MAOA males to aversive events (e.g. abuse) during childhood. The combination of a low-MAOA genotype with such events seem to produce abnormal regulation (through the cingulate) of the amygdala and an increased predisposition to impulsivity and violence. As the authors note:

Predisposition to impulsive violence by means of abnormal activation and regulation of emotion-related amygdala function might be further enhanced by deficient neural systems for cognitive control, especially over inhibition, the capacity to suppress prepotent but inappropriate behavior that might originate from a dysregulated affective response. Although the rostral cingulate is key to the regulation of acute affective arousal and emotional learning, inhibitory control of prepotent cognitive responses is thought to be critically dependent on caudal aspects of anterior cingulate. Our study of genetic influences on cognitive impulse control revealed a sex-dependent impairment in precisely this area of cingulate, affecting men only. Our finding of a genotype-by-sex interaction in this region therefore provides a plausible neural mechanism for reduced cognitive inhibitory control in risk allele-carrying males, suggesting synergistic impairment in cognitive and emotional neural regulatory mechanisms that might render MAOA-L men at especially high risk for a neural phenotype that plausibly relates to the slightly greater probability of impulsive violence.

Endnote: it might be useful to note that this study was conducted on healthy, non-criminal volunteers. The obvious step next is to study crime offenders (different types) and the complex interplay between genes, gender and childhood events.

-Thomas

Read Full Post »

Did you know?

The main excitatory neurotransmitter in the central nervous system in mammals is glutamate. In insects it's acetylcholine. Does this matter? Here's an answer from Gilles Laurent, in "Shall we even understand the fly's brain?" This is from the excellent book "23 problems in system neuroscience", as I have mentioned earlier.

"No, because either mechanism provides equivalent means for postsynaptic excitation over many timescales, using ionotropic and metabotropic receptors. The flow of activity within and across networks are, as far as we know, similar with either combination of neurotransmitter and receptors. On the other hand, the fact thatglutamate ended up being used in the mammalian CNS enabled the exploitation of receptor subtypes with interesting nonlinear properties (e.g. NMDA receptors) for a variety of fundamental tasks such as rythm generation or learning."

-Thomas

Read Full Post »

CIMBI alive

The Center for Integrated Molecular Brain Imaging (CIMBI) is now officially opened. The overall idea behind this massive project is to study cognitive, psychological and biological phenomena with a multi-modal approach, combining data from genotyping, PET scanning and MRI scanning. The main project of CIMBI is to study “the neural bases of personality dimensions that predispose individuals to affective and substance use disorders, with special emphasis on the serotonergic neurotransmitter system”. In other words: to study the biological mechanisms behind personality formation. They are currently recruiting (and looking for) the best-qualified personnel for the new available positions.

One part of the CIMBI project involves looking at how genes coding for seretonin affect the seretonin transport function, and furthermore how the function of seretonergic areas of the brain operate depending on the genetic makeup of a subject. In this latter part, I am involved in doing the MRI study, including three fMRI protocols:

1. Processing of facial affect – how genes affect the processing of facial expression, especially the difference between aversive and neutral faces.
2. Memory processing in the medial temporal lobe (MTL) – how different parts of the MTL make different contribution to specific phases in memory processing: preparation, encoding, rehearsal and retrieval.
3. Categorization task – the difference between choosing between high-specificity options (within-category choices, e.g. “donkey or zebra”) or low-specificity options (between-category choices, e.g. “living or non-living”)

Data will be combined between fMRI (BOLD and perfusion), genotype and seretonine function as measured with PET. In addition we are looking at the relative contribution of changes in volume and form of MTL areas to the overall signal differences found in other modalities.

Read Full Post »