Lecture_09_Decision Making

Question 1

Decision Making

Answer 1

Assessing and making choices between various OPTIONS
- Evaluated on their CONSEQUENCES

Question 2

Judgment

Answer 2

Estimating the PROBABILITY of various EVENTS
- Evaluated on their accuracy
- E.g. “Which one is a more likely cause of death – car accident or lung cancer?”

Question 3

Normative Theories

Answer 3

How people SHOULD make decisions

Question 4

Descriptive Theories

Answer 4

How people ACTUALLY make decisions

Question 5

Ideal, Rational Decisions

Answer 5

Based on Economics
- Make decisions based on logical reasoning
- Calculate and choose the option that maximizes expected value

Question 6

Expected Value

Answer 6

Expected value = (value of a given outcome) x (probability of that outcome)

Question 7

Which one is an optimal decision?
A) 50% chance of gaining 50 baht
B) 10% chance of gaining 100 baht

Answer 7

A) E(X) = (0.5) x 50 = 25 baht
B) E(X) = (0.1) x 100 = 10 baht
- So, the “optimal” decision is A), because it has a greater expected value
- Prospect theory

Question 8

Subjective Value

Answer 8

Based on Psychology
-The same option can have different values!
- Depends on the person and the situation (subjective value)
- Calculate and choose the option that maximizes Expected utility
- For that particular person, in that particular situation

Question 9

Expected Utility

Answer 9

Expected utility = (utility of a given outcome) x (probability of that outcome)

Question 10

Which one is the rational choice?
A) One KFC set for 100 baht
B) Fifty KFC sets on sale for 4,000 baht

Answer 10

B) is more rational but, with subjective value, people may choose A) more

Question 11

Principle of transitivity in rational decision making

Answer 11

• If A > B and B > C
• Then A > C

Question 12

Predictably Irrational Decisions

Answer 12

People don’t always choose the “optimal”, logical choice…
- Not random choices
- Consistent, but irrational (sub-optimal) choices

Question 13

Decision Making Under Risk

Answer 13

Rationally, people should always choose the option that maximize chances of gains, and minimize chances of losses

Question 14

Prospect Theory

Answer 14

• People are much more sensitive to potential losses than potential gains (loss aversion)
• “Losses loom larger than gains”
• So, people are willing to take risks to avoid losses

Question 15

Heuristics

Answer 15

• Framing bias/effect
• Availability heuristic
• Representativeness heuristic

Question 16

Framing Bias/Effect

Answer 16

Decisions can be influenced by SITUATIONAL ASPECTS that are irrelevant to optimal decision-making
- E.g., wording of the problem
- Positive vs. negative frames (but same outcome)

Question 17

Availability Heuristic

Answer 17

When judging the frequency of events, we tend to judge an event as more frequent if it is EASIER TO RETRIEVE specific examples from memory
- E.g. the more likely cause of death

Question 18

Representativeness Heuristic

Answer 18

• Compare how similar the example is to the category, rather than relying on valid statistical information.
• Stereotypes

Question 19

Brian is 32. He is intelligent, unimaginative, and somewhat lifeless. In school, he was strong in math but weak in languages and art.
Which of these is more likely?
A) Brian plays in a rock band.
B) Brian plays in a rock band and is an astrophysicist

Answer 19

A) is more likely
Because conjunction probabilities can never be higher than one of the constituent probabilities.
P(A and B) = P(A) x (P(B)
- Representativeness Heuristic

Question 20

Anchoring & Adjustment Heuristic

Answer 20

When calculating the value of something
- we start with an INITIAL GUESS (anchor)
- then ADJUST upward or downward
- But, adjustments are usually insufficient!
- Final estimates tend to remain close to the initial anchor

Question 21

Prior Beliefs Approaches

Answer 21

Our prior beliefs strongly influence the choices we make
- Combine current information with prior information to make the best
guess about a future outcome
- Usually works, but sometimes leads to over- or under-estimations
- E.g. today’s temp. < 20 c but it’s Thailand so didn’t bring sweater

Question 22

Dual Process Model: System 1

Answer 22

• Fast
• Associative reasoning
• Relies on prior knowledge
• Most of the time, very effective
• But in complex situations, may lead to “wrong” or biased choices.

Question 23

Dual Process Model: System 2

Answer 23

• Slow
• Rules of logic
• Mathematical calculations
• Analytical
• But may not always override a faulty System 1 judgment !

Question 24

Neuroeconomics (Decision Neuroscience)

Answer 24

Economics + Psychology + Neurobiology
To explain and predict the actual choices people make
- Explain observed choice, through
MATHEMATICAL models
- Explain MENTAL states that produce a choice, through conducting experiments
- Identify NEURAL mechanisms and evolutionary pressures that guide choices

Question 25

Decision-Making & AI

Answer 25

• Complex, unstructured problems
• Rapidly changing
• Flexible and adaptable
• Tailored to decision-making approach of the user