October 16, 2004
Brain Battles Between Short Term Emotions And Long Term Logic

When you give in to the impulse for an immediate reward over a longer term greater reward your emotional regions of your brain has beat out your logical regions.

You walk into a room and spy a plate of doughnuts dripping with chocolate frosting. But wait: You were saving your sweets allotment for a party later today. If it feels like one part of your brain is battling another, it probably is, according to a newly published study.

Researchers at four universities found two areas of the brain that appear to compete for control over behavior when a person attempts to balance near-term rewards with long-term goals. The research involved imaging people's brains as they made choices between small but immediate rewards or larger awards that they would receive later. The study grew out of the emerging discipline of neuroeconomics, which investigates the mental and neural processes that drive economic decision-making.

The study was a collaboration between Jonathan Cohen and Samuel McClure at Princeton's Center for the Study of Brain Mind and Behavior; David Laibson, professor of economics at Harvard University; and George Loewenstein, professor of economics and psychology at Carnegie Mellon University. Their study appears in the Oct. 15 issue of Science.

"This is part of a series of studies we've done that illustrate that we are rarely of one mind," said Cohen, also a faculty member at the University of Pittsburgh. "We have different neural systems that evolved to solve different types of problems, and our behavior is dictated by the competition or cooperation between them."

The researchers examined a much-studied economic dilemma in which consumers behave impatiently today but prefer/plan to act patiently in the future. For example, people who are offered the choice of $10 today or $11 tomorrow are likely choose to receive the lesser amount immediately. But if given a choice between $10 in one year or $11 in a year and a day, people often choose the higher, delayed amount.

In classic economic theory, this choice is irrational because people are inconsistent in their treatment of the day-long time delay. Until now, the cause of this pattern was unclear, with some arguing that the brain has a single decision-making process with a built-in inconsistency, and others, including the authors of the Science paper, arguing that the pattern results from the competing influence of two brain systems.

The researchers studied 14 Princeton University students who were asked to consider delayed reward problems while undergoing functional magnetic resonance imaging (fMRI), a procedure that shows what parts of the brain are active at all times. The students were offered choices between Amazon.com gift certificates ranging from $5 to $40 in value and larger amounts that could be obtained only by waiting some period, from two weeks to six weeks.

The study showed that decisions involving the possibility of immediate reward activated parts of the brain influenced heavily by brain systems that are associated with emotion. In contrast, all the decisions the students made -- whether short- or long-term -- activated brain systems that are associated with abstract reasoning.

Most important, when students had the choice of an immediate reward but chose the delayed option, the calculating regions of their brains were more strongly activated than their emotion systems, whereas when they chose the immediate reward, the activity of the two areas was comparable, with a slight trend toward more activity in the emotion system.

The researchers concluded that impulsive choices or preferences for short-term rewards result from the emotion-related parts of the brain winning out over the abstract-reasoning parts. "There are two different brain systems and one of them kicks in as you get really proximate to the reward," McClure said.

What does it say about economists that today it is a growing view among economists that factors other than pure reasoning often drive people's decisions? They are just figuring this out? They didn't understand this, say, 20 years ago?

The finding supports the growing view among economists that psychological factors other than pure reasoning often drive people's decisions.

Do smarter people have a greater ability to control their emotions? Are brains that have more gray matter dedicated to abstract reasoning also brains whose abstract reasoning areas are better able to win out in a fight with the emotional areas? Or do the emotional areas typically scale up as much in size as the abstract reasoning areas in people who have larger abstract reasoning areas and bigger brains?

"Our emotional brain has a hard time imagining the future, even though our logical brain clearly sees the future consequences of our current actions," Laibson said. "Our emotional brain wants to max out the credit card, order dessert and smoke a cigarette. Our logical brain knows we should save for retirement, go for a jog and quit smoking. To understand why we feel internally conflicted, it will help to know how myopic and forward-looking brain systems value rewards and how these systems talk to one another."

The findings also may cast light on other forms of impulsive behavior and drug addiction.

This result explains what Spock had to do to become more Vulcan: He had to suppress the human dopaminergic circuits in his brain that caused him to feel emotions. But pure-blooded Vulcans have an easier job of mastering their emotions. The Vulcans must have either genetically engineered their species to down-regulate emotion producing areas of the brain or they engaged in generations of eugenic breeding practices ("Match maker, match maker, make me a logical match, catch me an unemotional catch") aimed at suppressing impulsiveness and emotional reactions to immediate stimuli. Certainly, as Sarek demonstrated, the Vulcans are not above the use of sophisticated reproductive technology to produce unusual offspring. Just because the Vulcans never have admitted to genetically engineering their species in any TV show episode that doesn't mean they didn't do so. After all, they hid Pon Farr as long as they could. They obviously have their secrets.

Dopaminergic circuits (neurons that release and accept dopamine neurotransmitter molecules) make us do what we ought not do.

"Our results help explain how and why a wide range of situations that produce emotional reactions, such as the sight, touch or smell of a desirable object, often cause people to take impulsive actions that they later regret," Loewenstein said. Such psychological cues are known to trigger dopamine-related circuits in the brain similar to the ones that responded to immediate rewards in the current study.

Concerning addiction, said Loewenstein, the findings help explain some aspects of the problem, such as why addicts become so focused on immediate gratification when they are craving a drug. The dopamine-related brain areas that dominated short-term choices among the study subjects also are known to be activated when addicts are craving drugs.

The researchers are now trying to pin down what kinds of rewards and how short a delay are needed to trigger the dopamine-related reaction. Their ultimate goal is to better understand how the emotion-related and calculating systems interact and to understand how the brain governs which system comes out victorious.

These results also suggest that Vulcans may be resistant to drug addiction. Vulcan neural physiology probably doesn't have dopaminergic neurons capable of responding much to opiates for example.

It would be interesting to take a large population of youths, test their tendency to choose smaller shorter term rewards over larger longer term rewards, and then follow them in a longitudinal study to see if the youths that have a stronger preference for shorter term rewards are more likely to become cigarette smokers, alcoholics, drug addicts, and convicted criminals.

It would also be interesting to test tendency toward choosing smaller shorter term versus larger longer term rewards as a function of IQ. Are people with 100 IQs more likely to go for the immediate reward than people with 130 IQs? Anyone know if that sort of experiment has been done? Or is IQ research too much of a taboo subject for that kind of work to be done by economists?

Share |      Randall Parker, 2004 October 16 12:49 PM  Brain Economics

Patri Friedman said at October 16, 2004 1:26 PM:

An interesting book on the subject of short-term, automatic responses vs. long-term reasoned ones is The Robot's Rebellion. Among other things, it argues that increasing rationality (doing what you think you should) will have a much greater impact on society than increasing intelligence. Making behavioral economics-type mistakes happens far more often than not being smart enough to figure something out.

It would certainly be interesting to study impulsiveness vs. IQ. Also to compare impulsiveness to success in various arenas - is self-control a better predictor of income than IQ?

SpakKadi said at October 16, 2004 2:19 PM:

I don’t know that IQ would necessarily correlate with the ability to control impulsive decisions. Even if the logical side of the brain is strong, the emotional side of the brain may still be strong enough to overpower the logical side. For instance, if someone is obsessive compulsive, they may be very intelligent but still have difficulty controlling their impulses.

However, I would be interested to see the correlation between impulse control and success vs IQ and success. People of average intelligence who are willing to work hard for a big payoff later can be more successful than a genius who thinks only of immediate needs. Of course, the next question is, is it possible to think so long term that you sacrifice too much in the short term? In other words, is it necessarily a good thing for logic to always win out? (Spock’s trouble dealing with humans would be an example)

Jody said at October 16, 2004 2:22 PM:

So Freud was right? Sure sounds like id vs super ego to me.

Randall Parker said at October 16, 2004 2:35 PM:

Jody, Yes, the thought of the id and superego crossed my mind as well. But I haven't read Freud and so it is not clear to me that he really attributed to the id and superego the same attributes as these neuroscientists are attributing to the emotional and rational brains. Maybe. Maybe not.

Randall Parker said at October 16, 2004 3:21 PM:


IQ is very predictive of economic achievements. See, for example, this posting from another blogger on IQ and economic success for children from the same family.

Obviously, IQ doesn't explain everything. One's degree of impulsiveness has got to be important as well. But there is just no way that a 100 IQ person with little impulsiveness is going to get rich by writing a really complex computer program or a new computer chip. You need the raw intellectual octane to achieve in many fields. The psychometric research bears this out.

Garson Poole said at October 16, 2004 3:29 PM:

In the cartoon "Popeye" a plump character named J. Wellington Wimpy often proclaimed this classic line: "I would gladly pay you Tuesday for a hamburger today." The humor in the Wimpy character was provided by the universal human recognition of deception regarding delayed payments. When I originally read about the experiment that "offered the choice of $10 today or $11 tomorrow" I thought that the experimenters would surely take into account the possibility that subjects would assign a probabilistic weight to deception.

Experiments conducted on college students are often filled with subterfuge, and the subjects are typically aware of these manipulative scenarios. Imagine a student accepts the offer of $11 dollars to be paid tomorrow. When the student arrives the next day he or she is placed in the functional magnetic resonance imaging device and told that the payment will not be given. The researchers then examine the fMRI data to learn about brain states associated with human anger. A subject may not envision this precise scenario but he or she would be aware that the promised money might never be forthcoming.

Instead of the crudely didactic interpretation of the experimental results provided in the article there is an alternative. Suppose that the "brain systems that are associated with emotion" are also the parts of the brain that help to evaluate the deceptiveness of a situation. Perhaps the experimental subject places higher value on a reward of $10 today than $11 tomorrow because he or she assigns the immediate reward a substantially greater certainty. A “rational” calculation that assigns a probabilistic weight to deception might determine that the $10 reward is best.

What about explaining the other experimental result in which "given a choice between $10 in one year or $11 in a year and a day, people often choose the higher, delayed amount"? Here the evaluation of deception probabilities would differ. The experimental subject might think the following: If the money giver is actually reliable enough to give $10 in one year then the giver would also be reliable enough to give $11 in one year. This conclusion would be based on a complex model for the human psychology of other beings honed by experience.

In conclusion, "neuroeconomics" is a valuable new field to explore but perhaps researchers should try to avoid propounding overly simplistic or naive models of human thought.

Fly said at October 16, 2004 3:52 PM:

Clinton is fairly intelligent and yet demonstrated poor impulse control.

(Not that one anecdote means much.)

Fly said at October 16, 2004 3:58 PM:

Garson Poole, I had the same thought when I read the $10 vs. $11 example. The later description of the experiment showed they used a range of rewards and time delays for gratification and matched decision making with brain activity. I believe they’ve measured a real affect.

Fly said at October 16, 2004 4:00 PM:

Darn I was affected by the effect. Sigh.

Perry Willis said at October 17, 2004 8:01 AM:

Emotions have always been a part of economic theory. The power of incentives, of personal self-interest versus group interests, and the idea of time-preference, have long been a part of economics. Only now we are able to see why some of these things work the way they do in the brain.

Michael Vassar said at October 17, 2004 8:37 AM:

This would be more id vs ego. Superego is also non-rational. Id is basically non-rational biological drives, superego is non-rational socialization (note, neither is irrational) ego is wanna-be rational internal monologue. Puny, ephemeral, impotent regarding behavior except at the moment, but able to form habits by focusing on the moment and on shaping the superego.

Invisible Scientist said at October 17, 2004 11:26 AM:


I must add that even though when the logical side of my brain was clearly telling me
that I was being manipulated or conned by people, the fears from the emotional side often
forced me to censor that observation, and like sheep I went to my slaughter again and
again (often by the same people, too!)
This differs a little bit from the "impulsive behavior", but it is related somehow to the
same phenomenon.

Note that Richard Feynman's IQ was only 137, but he had a lot of energy and aggressiveness
to get things done.

Rob Sperry said at October 18, 2004 8:19 PM:

"Note that Richard Feynman's IQ was only 137, but he had a lot of energy and aggressiveness
to get things done."

It would be better to say that on a particular test he scored 137. He also high scored the putnum exam while finishing early...what test are you going to believe? He probably had a low tolerance for taking the test, combined with less than steller verbal ability (when he was young) relative to his mathematical skill. How do you test the IQ of someone much smarter than the people making the test?

Invisible Scientist said at October 19, 2004 4:41 AM:

Of course, Richard Feynman was probably much smarter than the people who have designed the
test, but all I was trying to say was that a high IQ is not enough, and that in addition to IQ,
some kind of motivation is necessary. I have met people with IQ at least 160, and for some
reason, some of these people were not all that outstanding, either due to imaginary
defects and other issues, or else they were valueing superficial quests instead of
focusing on a career that is original or creative. Fenynman had VERY strong enthusiasm. It turns
out that his father Melville really took good care of him, in the sense
that without invading the privacy of his son, he nurtured him by surrounding him with ideas
and toys that will stimulate the mind of his son. When he took his son to the zoo, he said to him: "You
now know the names of all the birds, but you still know nothing about the birds."

Rob Sperry said at October 19, 2004 8:31 PM:

"It turns out that his father Melville really took good care of him"

Feynman's recollections of his time with his father and the way he encouraged him to think is one of the inspirations I hope to draw on as a father.

Another axis that I think is relevant is the pleasure one gets from thinking about things verses doing things. Many high IQ people are delighted to think about an idea, but quickly loose enthusiasm when action is required. This is probably a reason that they are good at thinking, but also an equal explanation why they get little done.

I think the high productivity people tend to get a lot of pleasure out of both activities and feel a bit empty if they are not doing both. Feynman in describing why he always took positions that required some teaching remarked that he always pitied the people at the Princeton Institute for Advanced Study because all they had to do was sit around and think all day.

wade said at October 22, 2004 12:52 PM:

The "Marshmallow Test" explores this - I read about it in the book Emotional Intelligence. Four year olds were given one marshmallow and left alone in a room for 15 minutes. They were told that if they didn't eat it until the researcher came back then they would get a second marshmallow. Some ate the first one immediately, others waited. The interesting thing is that 10 years later these same kids were reevaluated and it was found that those that delayed gratification and waited were better students, scored higher on standardized tests, had fewer problems. etc than those that ate the one immediately. More interesting was that the marshmallow test was a better indicator of subsequent higher test scores than the IQ tests the 4 year olds took.

sandeep said at February 22, 2005 6:03 AM:

U can not measure intelligence....it is not something like comparing two computers by taking in to consideration parameters like memory , lcock rate etc.....the functioning of brain is non-linear..and we can not measure anything non-linear on our linear scales like IQ tests etc etc....sometimes a "not so good" person comes up with something
so extra-ordinary that so called "geniuses" would have never thought about it!!!!

chuck schmuck said at April 30, 2005 4:18 PM:

Didn't Ted Bundy have a high IQ ?

chuck schmuck said at April 30, 2005 4:21 PM:

What about Hitler ? He used no drugs or alcohol, but Rosevelt had a few drinks everyday!
Wasn't Hitler looking at the long term of his goals?

judic said at May 10, 2005 1:29 PM:

Hitler didn't use drugs?
www.google.com - Hitler and amphetamines.

There are plenty of sites that show he did use drugs. Hitler abused amphetamines. Stop using anecdotal "evidence."

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©