[The Abstract from the research report in the journal Science is appended below
the news stories.]

* * *

http://www.nature.com/news/2005/051205/full/051205-10.html

Published online: 8 December 2005; | doi:10.1038/news051205-10

Heart rules head on risky calls
Uncertainty activates the brain's sentimental centres.

Roxanne Khamsi

Illustration Omitted:
Unknown risks provoke emotional responses. © Getty

When faced with uncertainty, people try to make the most logical decision, given
the facts available. But a brain-imaging study has found that, when tackling
these tricky decisions, the brain's emotional areas also spring into action.

Ming Hsu of the California Institute of Technology in Pasadena and his
colleagues compared volunteers' brain activity in two betting games. As the
volunteers played, the scientists watched changes in their brain activity using
functional magnetic resonance imaging.

In one game, researchers gave volunteers the chance to guess the colour of a
card drawn from a deck containing equal numbers of red and blue cards, and to
bet on whether they were right.

In the other game, the ratio of red to blue cards remained unknown. Players in
this game were less likely to put money on their guess. And there was a burst of
activity in their brain's emotion-processing centres, the amygdala and the
orbitofrontal cortex. The study is published in this week's Science1.

Playing it safe

The volunteers didn't know that the odds of them guessing right were actually
the same in both games. Because players in the second game could only say 'red'
or 'blue', their chances of betting correctly remained at 50%, whatever the
ratio of blue and red cards.

That may sound odd: but as long as people are unbiased toward red or blue, they
are equally likely to say either colour. And so even if the deck contains only
red cards they still have a 50% chance of guessing correctly.

The scientists found that in the second game, the emotional parts of volunteers'
brains typically sparked into life a few seconds before they made their choice -
perhaps producing cautious feelings that swayed the subject against betting. The
researchers believe their finding supports the idea that our brain wiring makes
us averse to ambiguous risks.

Patients with a damaged orbitofrontal cortex bet the same regardless of whether
they knew the risk or not. "A lot of these patients end up financially
destitute," says Ming. He speculates that such people lack the emotional
response to uncertainty, and the aversion to risk, of those with an intact
orbitofrontal cortex.

Aldo Rustichini, a decision-theory researcher at the University of Minnesota in
Minneapolis, says that the findings nicely illustrate how emotion gets activated
in tough situations. But he points out that the study found that risk also
increased activity in analytical regions of the brain: "Decision-making involves
both emotion and reasoning. The results in the paper point to both."

"It's important to try to understand how these two large networks we usually
attribute to emotion and reason are interacting," Rustichini adds.

Ming and his fellow researchers intend to investigate whether indirect
stimulation of the emotion-control centres can make a person more cautious. This
could help to establish a causal link between amygdala activity and guarded
decision-making, says Ming.

References

Hsu M., Bhatt M., Adolphs R., Tranel D., Camerer C., et al. Science, 310. 1680 -
83 Doi:10.1126/science.1115327 (2005).

©2005 Nature Publishing Group

* * *

http://www.sciencemag.org/cgi/content/full/310/5754/1624?maxtoshow=&HITS=10&hits\
=10&RESULTFORMAT=&fulltext=hsu&searchid=1134111059445_19926&stored_search=&FIRST\
INDEX=0&sortspec=date

Science 9 December 2005:
Vol. 310. no. 5754, pp. 1624 - 1625
DOI: 10.1126/science.1122179

Perspectives
NEUROSCIENCE:
Emotion and Reason in Making Decisions
Aldo Rustichini*

Two decks of 100 cards each are on a table. The one on the right has 50 red and
50 blue cards; the one on the left also has red and blue cards, but I don't tell
you how many of each. You pick one card from each of the two decks, without
looking at the color. Now you can bet on the color of a card of your choosing
between the two, and I will give you $100 if you guess right. Do you bet on the
right or the left card? You may be inclined to choose the right. Before you
answer, note that sheer logic dictates that you should consider the left one as
just as good a bet too. Here is why. Choose the left deck, and flip a coin to
choose a color. If the card is red, you will match the color with 50%
probability. If it is blue, the same conclusion holds. Therefore, no matter what
the color of the left card, you have a 50-50 chance of winning $100, which is
also what the card on the right side will give you. Are you convinced? If not,
are you willing to give me an extra dollar to get your preferred choice?

Although the logic is impeccable, most people are not convinced and prefer to
bet on the right deck because "they know the probability." They are also willing
to pay to avoid the vagueness plaguing the left deck. If asked to pay, people
offer around $42 for the left deck and $45 for the right deck. The right deck's
average worth is $50; the $5 difference between the offer for the right deck and
its average worth is what economists call the risk premium, a way to measure
aversion to risk. The additional $3 difference between the prices offered for
the two decks is the ambiguity premium, a measure of aversion to the vagueness
of the probability. In real life, the ambiguity premium may be substantial. For
instance, it is a large part of the difference between the higher price of
stocks of domestic companies, as opposed to cheaper foreign ones: People like
better what they know.

Illustration Omitted:
Figure 1 Decisions, decisions. Our brain treats choices involving risk or
ambiguity differently. CREDIT: JOE SUTLIFF

Economists in recent decades have realized [since (1)] that people are averse to
ambiguity. To account for this behavior, they (2) have built and used a formal
decision theoretic model. It formulates the idea that when the probability is
not precise, people are inclined to consider the worst possible outcome of each
action they can take as the outcome that will occur. In our example, if you
choose the ambiguous card deck, the worst possible outcome for each color you
choose is $0. You are facing a malevolent opponent who can choose the outcome
that is least favorable to you.

This is now an accepted and widely applied model (3-6). But is this just a
clever mathematical model, or does it correspond to a real process in the brain?
This formal theoretic model of ambiguity aversion has two main predictions. The
first is that subjects approach a decision with ambiguous probabilities in the
same way as they do when they face a malevolent opponent. The second is that
they deal with this situation as a calculated risk: In choosing with ambiguous
probabilities, subjects estimate the worst case, how likely it is, and how much
it pays. Dealing with decisions facing ambiguity is a process involving both
emotion and reason. Is this what we observe?

On page 1680 of this issue, Hsu and colleagues (7) report on a functional
magnetic resonance imaging study that may give us physiological clues as to the
nature of ambiguity. The main result is that the brain treats the two card decks
in the example above in different ways. Distinct areas of the brain are active
when we evaluate ambiguous and risky choices. Moreover, patients with large
lesions that incorporate one of these areas (the orbitofrontal cortex) treat
ambiguous and risky choices differently from normal subjects.

Twenty-four different areas in the brain are more active under conditions of
ambiguity than risk. Among these regions, Hsu et al. focused on those that
previous researchers have, with some controversy, associated with the emotional
side of decision-making. However, a large number of these areas (located in the
temporal, parietal, and prefrontal lobes of the brain) deal with the estimation
of the values of the options, which suggests that the decision process
integrates emotional and computational components. The results confirm earlier
findings that not only are ambiguity and risk treated differently by the brain
(8), but so are related situations such as when one considers sure and risky
outcomes, or monetary gains and losses (9). Taken together, these findings
support the theory of ambiguity aversion that economists have described.

What is next? Elucidating the neural processes underlying decision-making may
help us understand important economic differences between ambiguity and risk.
Human attitude to risk fuels the substantial profits of two large business
sectors of our economy--gambling and insurance. In contrast, there is no sector
served specifically by our aversion to ambiguity. This difference between risk
and ambiguity is related to an experimental fact: If I ask you to choose
repeatedly among risky options, your risk premium remains stable. But recent
experimental evidence (10) suggests that the ambiguity premium declines as
subjects repeat their choices: People slowly adjust to ambiguity; they do not
adjust to risk. Just as we learn to act optimally given the actions of others
(the Nash equilibrium of game theory), by choosing repeatedly, one may be
learning, slowly, to deal with ambiguity in our choices.

References and Notes

1. D. Ellsberg, Q. J. Econ. 75, 643 (1961).
2. I. Gilboa, D. Schmeidler, J. Math. Econ. 18, 141 (1989). [Abstract]
3. L. G. Epstein, Am. Econ. Rev. 91, 45 (May 2001). [AER]
4. C. A. Sims, Am. Econ. Rev. 91, 51 (May 2001). [AER]
5. G. Chamberlain, Am. Econ. Rev. 91, 55 (May 2001). [AER]
6. L. P. Hansen, T. J. Sargent, Am. Econ. Rev. 91, 60 (May 2001). [AER]
7. M. Hsu, M. Bhatt, R. Adolphs, D. Trane, C. F. Camerer, Science 310, 1680
(2005).
8. J. Dickhaut et al., Proc. Natl. Acad. Sci. U.S.A. 100, 3536 (2003).
[Medline]
9. A. Rustichini, J. Dickhaut, P. Ghiradato, K. Smith, J. Pardo, Games Econ.
Behav. 52, 257 (2005). [Abstract]
10. J. Snell, I. Levy, A. Rustichini, P. W. Glimcher, paper presented at the
Society for Neuroscience meeting, Washington, DC, 12 to 16 November 2005.
11. Supported by NSF grant SES-0452477.


10.1126/science.1122179
The author is in the Department of Economics, University of Minnesota,
Minneapolis, MN 55455, USA. E-mail: arust@...

© 2005 American Association for the Advancement of Science. All Rights Reserved.

* * *

http://www.sciencemag.org/cgi/content/full/310/5754/1680?maxtoshow=&HITS=10&hits\
=10&RESULTFORMAT=&fulltext=hsu&searchid=1134111059445_19926&stored_search=&FIRST\
INDEX=0&sortspec=date

Science 9 December 2005:
Vol. 310. no. 5754, pp. 1680 - 1683
DOI: 10.1126/science.1115327

Reports
Neural Systems Responding to Degrees of Uncertainty in Human Decision-Making
Ming Hsu,1 Meghana Bhatt,1 Ralph Adolphs,1,2 Daniel Tranel,2 Colin F. Camerer1*

Much is known about how people make decisions under varying levels of
probability (risk). Less is known about the neural basis of decision-making when
probabilities are uncertain because of missing information (ambiguity). In
decision theory, ambiguity about probabilities should not affect choices. Using
functional brain imaging, we show that the level of ambiguity in choices
correlates positively with activation in the amygdala and orbitofrontal cortex,
and negatively with a striatal system. Moreover, striatal activity correlates
positively with expected reward. Neurological subjects with orbitofrontal
lesions were insensitive to the level of ambiguity and risk in behavioral
choices. These data suggest a general neural circuit responding to degrees of
uncertainty, contrary to decision theory.

1 Division of Humanities and Social Sciences, 228-77, California Institute of
Technology, Pasadena, CA 91125, USA.
2 University of Iowa Medical School, Iowa City, IA 52242, USA.

* To whom correspondence should be addressed. E-mail: camerer@...

*** NOTICE: In accordance with Title 17 U.S.C. Section 107, this material is
distributed, without profit, for research and educational purposes only. ***