Title: Decisions, Decisions: The Ellsberg Paradox and The Neural Foundations of Decision-Making under Uncertainty
1Decisions, DecisionsThe Ellsberg Paradox and
The Neural Foundations of Decision-Making
under Uncertainty
- Ming Hsu
- Everhart Lecture
2Simple Decisions Blackjack
3Simple Decisions Blackjack
4More Complicated Investing
Stock? Bond? Domestic? Foreign?
Diversify Think long-term
5Complicated Love/Marriage
Whether? Who? When? Where?
37 Rule (Mosteller, 1987) Dozen Rule (Todd,
1997)
6Little knowledge of probabilities
Precise knowledge of probabilities
7Uncertainty about uncertainty?
8Ellsberg Paradox
1961
9Urn I Risk
Most people indifferent between betting on red
versus blue
10Urn II Ambiguity
?
?
?
?
?
?
?
?
?
?
Most people indifferent between betting on red
versus blue
11Choose Between Urns
Urn II (Ambiguous)
Urn I (Risk)
?
?
?
?
?
?
?
?
?
?
Many people prefer betting on Urn I over Urn II.
12Where Is The Paradox?
sadly but persistently, having looked into
their hearts, found conflict with the axioms and
decided to satisfy their preferences and let
the axioms satisfy themselves. --Daniel
Ellsberg, Quarterly Journal of Economics (1961)
13Ellsberg Paradox
Urn II (Ambiguous)
Urn I (Risk)
- P(RedII)P(BlueII)
- P(RedII) lt 0.5
- P(BlueII) lt 0.5
P(RedI) P(BlueI) P(RedI) 0.5 P(BlueI) 0.5
?
?
?
?
?
?
?
?
?
?
P(RedI) P(BlueI) 1 P(RedII) P(BlueII) 1
14Verizon
Jennifer
or Deutsche Telekom
or Angelina
15Verizon or Deutsche Telecom?
French Poterba, American Economic Review (1991).
16Explaining Ambiguity Aversion
Like physicists, economists like laws of
nature (Law of Demand, Walras Law, etc.)
Murphys Law If anything can go wrong, it will.
People consider the worst possible outcome of
each action.
17Explaining Ambiguity Aversion
18Explaining Ambiguity Aversion
P(RedIIBetRed) 0 P(BlueIIBetBlue) 0
P(RedI) 0.5 P(BlueI) 0.5
19What Are We Missing?
Gilboa Schmeidlers model is a model of
ambiguity aversion. There are a number of other
models of ambiguity aversion.
Unanswered
Do these models really reflect actual
decision-making process? How are the relevant
variables interpreted and choices produced? Look
in the brain.
20The Bigger Picture
Economics formal, axiomatic, global.
Human Behavior
Psychology intuitive, empirical, local.
Neuroscience biological, circuitry, evolutionary.
21The Bigger Picture
Economics formal, axiomatic, global.
Neuroeconomics A mechanistic, behavioral, and
mathematical explanation of choice that
transcends each field separately. - Glimcher
and Rustichini. Science (2004)
Human Behavior
Psychology intuitive, empirical, local.
Neuroscience biological, circuitry, evolutionary.
22The Story of Phineas Gage
Cavendish, Vermont (September 13, 1848)
23The Story of Phineas Gage
fitful, irreverent, indulging at times in the
grossest profanity... -- Gages physician
- Impulsiveness
- Poor insight
- Impaired decision-making
- Both social and financial
24Fiorillo, Tobler, and Schultz. Science. (2003)
25Fiorillo, Tobler, and Schultz. Science. (2003)
26Fiorillo, Tobler, and Schultz. Science. (2003)
27Tools That We Used
Functional Magnetic Resonance Imaging (fMRI)
Brain Lesion Patients
28MRI Magnetization of Tissue
29fMRI Changes in Magnetization
Basal State
30fMRI Time Series Data
intensity
Time
voxel time series
31Statistical Modeling of fMRI Data
Time
Intensity
32Random Effects/Hierarchical Models
pdf
?21
Subj. 1
0
?1
Distribution of population effect
?2Pop
?Pop
33fMRI Experiment
Hsu, Bhatt, Adolphs, Tranel, and Camerer.
Science. (2005)
34fMRI Experiment
Hsu, Bhatt, Adolphs, Tranel, and Camerer.
Science. (2005)
35fMRI Experiment
Hsu, Bhatt, Adolphs, Tranel, and Camerer.
Science. (2005)
36Expected Reward Region
y - Brain response A(.) - Ambiguity
trials R(.) - Risk trials E(.) - Expected value
of choices W(.) - Nuisance parameters
37Lower Activity under Ambiguity
38Lower Activity under Ambiguity
Signal Change
39Region Reacting to Uncertainty
y - Brain response A(.) - Ambiguity
trials R(.) - Risk trials E(.) - Expected value
of choices W(.) - Nuisance parameters
N.B. This region does not correlate with expected
reward.
40Link Between Brain and Behavior
Brain Imaging Data
Behavioral Choice Data
41A Signal for Uncertainty?
?
Late
Early
42Lesion Subjects
43Lesion Experiment
100 Cards 50 Red 50 Black
100 Cards x Red 100-x Black
Choose between gamble worth 100 points OR Sure
payoffs of 15, 25, 30, 40 and 60 points.
44Lesion Patient Behavioral Data
45Estimated Risk and Ambiguity Attitudes
Orbitofrontal lesion patients more rational!
46Linking Neural, Behavioral, and Lesion Data
Brain Imaging Data
Behavioral Choice Data
47What have we learned?
48One System, Not Two
Signal Change
49Reward Value of Ambiguous Gambles
50Signal for Uncertainty
51No OFC ? No Ambiguity/Risk Aversion
52Where are we going?
53Neural Circuitry
?
54The Brain and Home Bias
55Why Ambiguity Averse?
he was a gambler at heartand assumed that he
could always beat the odds.
On Jeffrey Skilling From Bethany McLean and
Peter Elkind, Smartest Guys in the Room (2003).
56Acknowledgements
Colin Camerer Ralph Adolphs Daniel Tranel Steve
Quartz Peter Bossaerts
Meghana Bhatt Cédric Anen Shreesh Mysore
ELS Committee
57END