Lecture 7

Question 1

What is crisis-conflict?

Answer 1

2 subcomponents of a decision conflict e.g cost and fun

Question 2

What is Opportunity cost?

Answer 2

• Once you’ve chosen to do one thing, lost all the other things you could have been doing with that time

Question 3

What is an easy example of decision making?

Answer 3

• Chess: Goal is to win, options are various and choice are various, it forms an assessment of how good a choice is = probabilistic = picking highest expected utility
• Consequence = win/lose the game

Question 4

What is a hard example of decision making? Career choice

Answer 4

• You have goals, biases, options and decision to decide
• Choice among options to maximise expected utility via states of world: mutually exclusive and exhaustive
• Outcome is every combination of option and state
• Outcome of expected utilities = outcome utilities multiplied by state probabilities
• Applying the whole framework is challenging = so simplify it

Question 5

How do we simplify the framework?

Answer 5

• Goals: simplify e.g money and enjoyment = can be simplified
• Decision to decide
• Biases: how can you be distorting info, and heuristics
• Options: 2+ possible behaviours unless multistop analysis needed
• Choice
• States: reduce
• Outcomes
• Outcome utilises
• State probabilities
• Consequences
• Evaluations

Question 6

What are the types of decision frameworks?

Answer 6

• Pros and cons
• GOO: goals, outcomes, options
• Weighted pros/cons
• Decision Trees
• ICE: importance weighted criterion evaluation
• MEU

Question 7

What are the pros and cons of the pros and cons decision framework?

Answer 7

• Could do a life decision with pros and cons where one thing is blatantly obvious
• Pros/cons are weighted the same even if one pro has more utility than a con
• Possible states given options are missing, unclearly defined and inappropriately unitary, vague probabilities
• Elimination by aspects: pick components of outcomes in order of their importance and eliminate options below a criterion value on each e.g to be a rockstar = need to sing well, therefore I cannot be, or I hate programming therefore no videogame designer

Question 8

What are the pros/cons of GOO?

Answer 8

• Structures info well
• Have a grid stating options, and relative components of outcomes
• Specifies goals too = not too specified, does not need to be mutually exclusive
• Then have to say who wins for each category
• Allows you to see conflict within goals
• Cannot eliminate by aspect
• BUT you can order your options
• Possible states of the world are merged and probabilities aren’t quantified = outcome uncertainty is not set
• Multiple criteria in utilities conflict: does not represent or revolve which is more important

Question 9

What are the pros/cons of weighted pros/cons?

Answer 9

• Use GOO and assign a quantitative sub-utility for each option an outcome
• Principle of dominance does not apply: no option wins on all three utility subcomponents
• Utility ambiguity e.g how does a + for enjoyment compare to a - for money
• Probabilities are still mostly ignored

Question 10

What are the pros/cons of decision trees?

Answer 10

• Utility only includes money as money can be a reflection of other things
• Same thing as full framework but utility is only composed of money

Question 11

What are the pros/cons of ICE?

Answer 11

• Distinction between a decision tree and ice if there’s only a single component to the utility e.g money, but the full framework has outcome uncertainty and multiple components to utility e.g money and enjoyment
• Use GOO and assign values to the table for importance weighted
• Ignores state probabilities
• Must specify states
• Ignores configural utility: configuration of components was a lot higher than all separate utilities added