Choices, Values, & Frames

Kahneman & Tversky APA Award Address

-we make literally thousands of decisions every day

-some are minor, such as

what to eat for breakfast
should I wear my light or heavy coat
which parking spot
where do I sit in class
deal or no deal

-others are much more consequential

should I stop to help a person in need
who should I vote for for president
should I have surgery
what job should I take

- we can make them knowingly( System 2; reflective) or unknowingly( System 1; intuitive)

Decision-making has both normative & descriptive processes

-Normative: concerned with rational and logical decision-making processes

-Descriptive: concerned with peoples decision-making as it is, not as they should be

I. Rational vs. Emotional Processes

A. Normative Models: assumes we try to make optimal decisions

Expected Utility Theory: von Neumann & Morgenstern

-proposes that that decision-making involves making a computation of utility, an indication of overall value (
expectation) for each possible outcome in a decision-making scenario

-we then consider all possible alternatives and rank alternatives in order of preference

-this lets us determine whether each alternative is more desirable, less desirable, or equally desirable, compared to each
competing one

-then we choose the most desirable one

Axioms of Rational Decision-Makers

1. Transitivity: If A is preferred to B; and B is preferred to C; then A should be preferred to C

2. Dominance: If A is at least as good as B in every respect and better than B in at least one respect, then A
should be preferred to B

3. Invariance : preference order between alternatives should not matter on order in which they are presented.

Problems:
-people routinely choose things that they are not happy with, not good for them
-we often don't have access to all relevant information
-we often make decisions quickly, without time needed to weigh all possibilities
-our thought processes are susceptible to bias
-we often decide on basis on emotional response, not rational process
B. Descriptive Models: focus on how people actually make decision
-try to account for tendencies humans have to misinterpret and misrepresent decision-making possibilities

Heuristics in Decision Making: Kahneman & Tversky

-mental shortcuts or rule of thumb that we typically use
-often occurs at unconscious level
-allows us to save conscious thought for other purposes
-can be adaptive because they allow for quick, reasonably good decisions, rather than weighing all the evidence

Problems: Can result in specific biases, which may lead to errors or faulty decisions

Examples:
1. Availability Heuristic: assess the likelihood of risks by asking how readily examples come to mind.
- Death estimates: if people can think of a vivid and relevant experience more likely to be concerned and rate the
probability as high.
Homicides are more available, so people tend to believe that more people die from them than from suicides
Tornadoes are more available than asthma attacks, etc

2. Representativeness Heuristic: similarity to our stereotype

3. Escalation of Commitment:
Let's say your elderly dad has a beloved car. Its reliability
was legendary, but it has started to have problems. He gets
one thing fixed, and something else goes wrong. Each fix
doesn't cost much, but they add up, and then the problems
start to get bigger. Your dad is convinced the next repair will get the car as good as new.
Would you advise him to pull the plug and get rid of the car?

Or consider this. A friend invests some money after getting a tip about a stock. The price soars, and your friend gains
10 percent overnight. He immediately doubles his investment. A week later, the thing tanks, and
he is in the red. A month later, it dives again, and he has lost a quarter of his investment.
Should he cut his losses and sell?

A woman you care about falls in love. After many years of a

happy relationship, the person she is with develops a vicious streak, starts smashing things and occasionally gives her a
black eye.

Would you tell her to walk out of the relationship?

4. Affective Forecasting: Expectation for how a decision will affect us in the future is powerful force in decision-
making.

-However, we are not good at predicting how we will feel in future

-After being diagnosed with serious illness, breaking up with partner, moving to new state, taking new job.

5. Optimism Bias:
-results in inaccurate self-assesment.

Second marriage is the triumph of hope over experience

II. Decision-making under Conditions of Risk

-made without advance knowledge of consequences

A. Risk Aversion Vs. Risk-Seeking
- Bernoulli observed that people are generally averse to risk
-risk aversion decreases with increasing weath

Example: Raffle for $1000. You have 85 tickets, out of 100 total. (85% Chance you will win; 15% chance you will
lose.)

Monty Hall comes up to you and offers you $800 to walk away from raffle. (Sure thing)

Result: Most people will pick the sure thing, over the gamble.

Yet, expectation of gain is higher with the raffle.

Expectation: weighted average. where each possible outcome is weighted by the probability of occurrence.

Raffle: .85 X $1000 + .15 X $.0 = $850

Sure Thing: 1.0 X $800= $800

Note: Rational choice would be choosing raffle, based on expected utility or normative model, yet most people pick the
sure thing.

Risk Aversion: preference for sure gain, over chance of bigger, but uncertain, return.

Do we always pick the sure thing? Not if the sure thing is a loss, then we are more likely to gamble on a potentially
smaller loss, but bigger risk.

Risk Seeking: Faced with a sure loss of $800. Given chance to pick situation with 85% chance to lose $1,000 ( but
15% chance to lose nothing).

Most people pick gamble, yet expectation of gamble ( -$850) is greater than expectation of sure loss (-$800).

General rule: risk averse when gains are involved; risk-seeking when losses are involved.

What determines our choice? Is it amount of money involved?

Bernouilli says it is SUBJECTIVE VALUE-

Size of a gain ( or loss) is what is important: the difference between a loss of $200 and a loss of $ 100 is perceived as
greater than a loss of $1,200 and a loss of $1,100.

Figure 1: Value Function.

Plots losses and gains vs. value.

-S-shaped function

Loss Aversion: function is steeper for losses than for gains ( loss of $X is more aversive than gain of $X is attractive)

How much are you willing to risk for possible gain of X

Most dont risk $10 for 50/50 chance of winning $10, 20, or $30. Need to have much higher value for potential gain
B. Framing Effects: risky outcomes are characterized by their possible outcomes and by the probabilities of those
outcomes. Same option can be framed or described in different ways.

-can be framed as gains or losses relative to status quo, or as asset positions that incorporate initial wealth. Shouldnt
make a difference based on order

Consider having to choose one of two options or, alternatively, having to reject
one of two options. Under the standard analysis of choice, the two tasks are
interchangeable. In a binary choice situation it should not matter whether people
are asked which option they prefer, or which they would reject. Because it is the
options themselves that are assumed to matter, not the way in which they are
described, if people prefer the first they will reject the second, and vice versa.

1. Invariance Axiom.

EXAMPLE: The US is preparing for the outbreak of a disease that will kill 600 people. Two alternative programs are
proposed. Scientific estimates of the consequences are as follows:

Scenario 1.
If Program A is adopted, 200 people will be saved

If Program B is adopted, One third probability that 600 will be saved and two-thirds probability that nobody will be
saved.

Which do you choose?

Results:
Most choose Program A ( 72%) , but both are equivalent.

A: Expectation 1.0 X 200 = 200

B: .33X 600 + .66 X 0 = 200

-Framed to emphasize reference state and two possible gains.. We pick the sure thing. Show Risk Aversion.

Scenario 2:
If Program C is chosen, 400 people will die
If Program D is chosen, one third probability nobody will die and two thirds probability that 600 people will die

Results:
most choose Program B ( 78%).

Framed to emphasize reference state with two possible losses. We pick the gamble where no one dies over the sure
death of 400. Show Risk-seeking. (NOTE application to gambling)

Yet in both scenarios, outcomes and probabilities are identical with exception that gains emphasized in first set of
alternatives, losses emphasized in second scenario.

2. Formulation Effects

process that controls framing of outcomes and events.

Formulation Effect: change of wording from lives saved to lives lost produces big change in preference from risk
aversion to risk seeking.

-passive acceptance of formulation given

Implications for Policy

a. Medical decision-making ( McNeil, et al, 1982; NEJM)

Scenario: patient is diagnosed with lung cancer; you must decide which of two treatments to use- surgery vs. radiation

Scenario 1: surgery has 10% mortality rate during treatment

Scenario 2: Surgery has 90% survival rate during treatment

If given scenario 1, subjects pick radiation more often. If given scenario 2, pick surgery more often
-mortality, emphasizes loss, triggers risk seeking; survival, emphasizes gain, triggers risk aversion.

b. Consumer decision-making: (Thaler, 1980)

-formulations can be manipulated for certain outcomes

labeling price differences for credit-card vs. cash purchases

Scenario 1: cash-discount

Scenario 2: credit-card surcharge

Lobbyists wanted it labeled as cash-discount. WHY?

-frames difference in price as gain or loss. Consumers less likely to accept a surcharge than to forgo a discount.

EX: University announced that faculty will increase contribution to retirement accounts, not university will decrease
salary

c. Social decision-making: Sharif ( 1993)

We propose that the positive features of options (their pros) will loom larger when choosing, whereas
the negative features of options (their cons) will be weighted more heavily when rejecting. It is natural to select an
option because of its positive features, and to reject an option because of its negative features. To the extent that people
base their decisions on reasons for and against the options under consideration, they are likely to focus on reasons for
choosing an option when deciding which to choose, and to focus on reasons for rejecting an option when deciding
which to reject.

Custody Example
Imagine that you serve on the jury of an only-child sole-custody case following

a relatively messy divorce. The facts of the case are complicated by ambiguous
economic, social, and emotional considerations, and you decide to base your
decision entirely on the following few observations.

To which parent would you award sole custody of the child?

Which parent would you deny sole custody of the child?]

Parent A: average income

average health
average working hours
reasonable rapport with the child
relatively stable social life
Award Deny
36% 45%

Parent B: above-average income

very close relationship with the child
extremely active social life
lots of work-related travel
minor health problems

Award: 64% Deny: 55%

Vacation Example
Prefer:
Imagine that you are planning a week vacation in a warm spot over spring
break. You currently have two options that are reasonably priced. The travel brochure gives only a limited amount of
information about the two options. Given the information available, which vacation spot would you prefer?

Cancel:
Imagine that you are planning a week vacation in a warm spot over spring
break. You currently have two options that are reasonably priced, but you can
no longer retain your reservation in both. The travel brochure gives only a
limited amount of information about the two options. Given the information
available, which reservation do you decide to cancel?

Spot A: average weather

average beaches
medium-quality hotel
medium-temperature water
average nightlife

Prefer Cancel
33% 52%

Spot B: lots of sunshine

gorgeous beaches and coral reefs
ultra-modern hotel
very cold water
very strong winds
no nightlife

Prefer Cancel
67% 48%

These results demonstrate that options are not simply

ordered according to value, with the more attractive selected and the less
attractive rejected. Instead, it appears that the relative importance of options
strengths and weaknesses varies with the nature of the task.

As a result, we are
significantly more likely to end up in spot B when we ask ourselves which we
prefer than when we contemplate which to cancel (67% vs. 52%, z = 2.83,
p < .OOl).
One of the most basic assumptions of the rational theory of choice is the
principle of procedure invariance, which requires strategically equivalent methods
of elication to yield identical preferences (see Tversky et al., 1988, for discussion).
The choose-reject discrepancy represents a predictable failure of procedure
invariance.

C. Changes or States: Prospect Theory

1. Psychophysics of chance

50% change to win $150

50% chance to lose $100

Would you accept this gamble?

Results: most people reject. Dont accept unless possible win is at least twice the size of the possible loss.
Attractiveness of possible gain is not nearly enough to compensate for aversiveness of possible loss. Need to greatly
increase possible gain to overcome risk-aversion.

2. Reference Dependence in Decision-making

Perception is reference dependent. Perceived attributes reflect the contrast between that stimulus and a context of prior
and concurrent stimuli.

Example: 3 buckets of water, cold, tepid, warm. Left hand in cold, right in warm. Adapt then plunge in tepid/.
Result??

Scenario 1: Imagine you are about to purchase a jacket for $125 and a calculator for $15. Calculator salesman
informs you that the calculator is on sale for $10 at other branch, 20 minutes away? Would you make trip to store?

Results: 68% say yes

Scenario 2: Imagine you are about to purchase a jacket for $15 and a calculator for $125. Calculator salesman
informs you that the calculator is on sale for $120 at other branch, 20 minutes away? Would you make trip to store?

Results: 29% say yes

But, in both cases, save $5. Reflects that decisions are not about absolute differences, but relative differences.

Scenario 3: You win $30 for sure.

Do you want to flip a coin to win or lose $9?

Results: 70% say yes

Scenario 4: You win $30 for sure.

Do you want to flip a coin to for either $21 or $39?

Results: 43% say yes

Change reference point.

3. Endowment Effects: reluctance of people to part from something they already have

-leads to preference for status quo

-leads to discrepancies in buying and selling prices ( violates standard economic theory)