Brain Chemical Underpins Social Interaction, And Why People Make Irrational Decisions

Other research shows that the release of the brain chemical serotonin, which plays a central role in clinical depression, is precisely tuned to various aspects of decision-making and reward-related behavior. New findings also show that the chemical has a significant role in maintaining our social networks, encouraging cooperation and anchoring relationships. These both are findings suggesting that the serotonin brain systems that go awry in depression normally play a critical role in supporting healthy and efficient decision-making.

"Over the course of the last few years, there has been an explosion in our understanding of how humans and animals make decisions," says Paul Glimcher, PhD, of New York University. "Ten years ago, we knew almost nothing about how the human brain weighed costs and benefits to arrive at a choice. Today, there are exciting new discoveries every year. These four studies are examples of just how fast our understanding is growing. And what these studies make clear is that insights from this kind of neuroeconomic research will influence both the structure of our future financial markets and clinical strategies we use to treat mental illness. That's a very cool combination."

New imaging work focuses on our aversion to loss, showing that choices we make when we face losses may rely much more on emotional brain systems than decisions that involve gains of equal, or even greater, size.

Working with 20 undergraduates, PhD student Peter Sokol-Hessner, at New York University, recorded participants' decisions when faced with choices representing various investing scenarios.

Subjects were asked a series of questions as their brain activity was monitored by functional magnetic resonance imaging (fMRI). Sokol-Hessner was able to correlate loss-averse behavior with activity in the amygdala, an area of the brain known to be involved in processing emotions-oftentimes, fear.

"Notably, in contrast to other research, these areas of correlation are not the same as those areas identified as generally active during valuation and decision-making relative to rest, such as the striatum and medial prefrontal cortex," says Sokol-Hessner.

Sokol-Hessner asked his subjects to make two series of 150 choices about how to spend $30. In both series, the subjects chose either to make a risky investing gamble or to settle on a guaranteed amount. For example, the decision might be between a 50-50 gamble in which the participant would either lose $16 or win $25, or the sure choice of $0 (neither winning nor losing a cent). In one series, the subjects evaluated each of the choices independently. For the second, researchers asked them to think "like a trader," evaluating each choice as one in a portfolio of investing decisions. In both cases, they learned the result of each decision immediately.

In a previous version of the study, the researchers had found that "subjects sweat significantly more, per dollar, to losses than gains," says Sokol-Hessner. "This 'over-arousal' correlated with behavioral loss aversion, suggesting a specific role for emotions in choice."

More recently, Sokol-Hessner confirmed results from previous studies showing that as a group, people fear a loss more than they value a gain of an equal amount, but his detailed results showed that, at the individual level, only half his subjects were loss-averse. Equally, about half were not, and some subjects even valued the gain more highly than dreading the loss. He also found substantial variation among his subjects in terms of how much risk they were willing to take and how consistent their decisions were.

Yet no matter their individual profile, Sokol-Hessner found that the subjects made choices that were less loss-averse when they thought about their choices as part of a portfolio. This was true whether or not subjects showed loss-averse tendencies in general.

"These findings are of interest because they shed light on some possible behavioral and neural differences between professional and amateur traders and suggest that the distance between the two can be reduced by something as simple as a cognitive strategy," Sokol-Hessner says.

For future research, Sokol-Hessner may recruit professional traders and compare the biological basis of their investment decision-making. "The integration of methods from economics, psychology, and neuroscience is a signature of neuroeconomic research," he says. "This kind of research has great promise to extend our understanding of how people make decisions and of how they can reliably alter the mechanisms of their own decision-making by taking alternate perspectives on the same choices."

The degree to which we are risk-averse also varies from person to person, as does our tolerance of ambiguity-for example, how much we would prefer a 50 percent chance of winning $20 over an unknown probability of winning $100. This is of particular interest to financial decision-makers because nearly all humans show an irrational aversion to ambiguous investments, even if those investments are likely to perform well. This is true even for individuals who are comfortable with risky investments. But new research using fMRI now indicates these two kinds of decisions in fact rely on activity in the same areas of the brain.

"We know very little about how idiosyncratic risk aversion and ambiguity aversion arise in the human brain," says Ifat Levy, PhD, of New York University. "Indeed, there is not yet even consensus about whether our fear of ambiguous situations reflects the activity of a dedicated brain system or simply a fine-tuning of the systems that represent risk.

"We suggest that neural activation in the basal ganglia and prefrontal cortex serves as a 'common neural currency' for valuing the many different kinds of opportunities we face as human decision-makers."

Working with 10 people, Levy showed her subjects images of jars with red or blue poker chips; they could tell the proportion of red chips in some jars but not others. By drawing a red poker chip, subjects could win money, but the percentage of red chips in the jar, how clear or cloudy the jar was, and the amount of money varied. For each trial, subjects pushed a button to indicate whether they chose to draw a chip or to play a second lottery, in which they had a 50 percent chance of winning $5.

After 360 trials over two sessions, six were played for real money. At the end of the experiment, Levy correlated the brain activity recorded by the imaging scans with objective as well as subjective parameters indicating each subject's preference for risk and ambiguity. Levy found that the same levels of risk-aversion and ambiguity-aversion matched certain patterns of activity in the medial part of the frontal cortex and the basal ganglia, areas known to play a role in decision-making.

"These results cast new light on a growing body of evidence that suggests our brains possess a single, central system for valuing the objects of our decisions in situations ranging from impulsive decision-making to ambiguous investing," says Levy.

"Based upon these data and others, we suggest that this network of brain areas generates the idiosyncratic valuations we place on the options before us, regardless of their nature or the contexts that influence that valuation.

"In other words," she says, "if activity in your prefrontal cortex is strongly affected by risk, ambiguity, delay of gratification, or loss, then that's the kind of person you are."

Recent research with animals shows that the release of the neurotransmitter serotonin is precisely tuned to various aspects and stages of reward-related behavior. Such results may provide a basis for developing more selective medications with fewer side effects for disorders such as depression, obsessive-compulsive disorder, and schizophrenia.

Serotonin acts as on-off switch, controlling various emotional states, and drugs that alter the action of serotonin have been used to treat depression and anxiety disorders for more than a decade. Focusing on the raphe nucleus, the brain region that controls serotonin release, Zachary Mainen, PhD, of Cold Spring Harbor Laboratory, trained rats to associate certain smells with a reward of water. Mainen will be presenting at Neuroscience 2007 at a minisymposium titled "Serotonin and Decision-Making."

On smelling an odor, the rats in the study would react to the scent, poking their noses into one of two holes. When the rat chose the correct hole, it received a drop of water as a reward. Throughout each experiment, comprising several hundred such decisions, Mainen monitored the activity of individual nerve cells in the dorsal raphe nucleus, an area located deep in the brain.

He found that separate subsets of raphe nucleus neurons responded independently to smelling, movement, and reward-related behaviors. In most cases, the nerve cells fired almost immediately-within tens of milliseconds.

Mainen also noticed that raphe nucleus nerve cells fired more when the animal had to ignore distracting sensory information, and stopped firing when it was concentrating on important sensory signals, underscoring a role for serotonin in the feedback loop that adjusts the brain's response to sensory stimulation in healthy individuals.

"Serotonin is a primary target for treatment of depression, anxiety, and other psychiatric disorders, but its function is not well understood," says Mainen. "It is considered particularly enigmatic because it seems to be involved in such a wide variety of brain and behavioral functions.

"These results suggest a specific cellular basis for the diversity of serotonin functions and possible avenues for development of more specific treatments for disorders such as major depression."

Future research will focus on serotonin release in response to more specific behavioral tasks and attempt to distinguish between nerve cells in the raphe nucleus that release serotonin and those that do not.

"These approaches will allow us to stimulate serotonin neurons artificially in order to test their influence on specific behaviors in animals," says Mainen. "In future studies, we would like to examine the impact of psychoactive drugs that target the serotonin system on the firing of different classes of serotonin neurons."

Other findings clarify the role of serotonin in decision-making within a social context: It may encourage cooperative behavior and help solidify social bonds by reinforcing the value we see in others.

Previous work has shown a link between depression and serotonin dysfunction and indicates a role for the neurotransmitter in behavior emphasizing an affiliation between people. Research based on a game called the prisoner's dilemma found that mutual cooperation enhances activity in brain circuits playing a role in reinforcement, indicating that cooperative behavior is rewarding in its own right.

Robert Rogers, PhD, at Oxford University, also used the prisoner's dilemma in his work, in which he altered serotonin levels and evaluated the effect of this alteration on participants' behavior. Rogers also will be speaking at the minisymposium.

In the prisoner's dilemma game, people make choices that affect each other, either favoring one person as a result of unequal sharing, or expressing more cooperative decision-making.

Working with subjects in pairs, Rogers blocked levels of l-tryptophan-the precursor of serotonin-in some participants. This had the effect of temporarily decreasing serotonin levels in these subjects.

As a result, Rogers found, the game participants became less willing to cooperate with each other. Lower serotonin levels also had the effect of changing the subjects' judgment of the social characteristics of others.

Rogers suggests that lower serotonin levels "diminished the reward value of cooperative behavior." Serotonin may also "play a role in modulating the cognitions that underpin dependable relationships with our social partners," he says.

Such findings also may indicate a role for prisoner's dilemma and other models based on game theory in enhancing understanding of and developing therapies for psychiatric disorders.