Video Module 28: Decision Making Flashcards
probability in decision making
how likely an outcome is
utility in decision making
how important an outcome is
- can be either subjective or objective
- can be influenced by psychological factors
utility theory
the theory of decision making which assumes that humans are rational actors who always choose the option that provides them with the most *utility
- In utility theory, utility is the “expected value” of a choice
—expected value = the product of the probability of an outcome and the value of an outcome
- e.g. You have a 1/20 chance of winning $100
—probability: 0.05
—value: $100
—expected value: 0.05x100 = $5
—utility in utility theory: $5
When are people likely to be more risk-averse?
People are more likely to be risk averse when presented with a definite gain versus a risky gain.
- e.g. Definitely win $50, or enter a 50% chance to win $100?
—According to utility theory, the utility of these two options is the same, therefore it should be a 50/50 split. However, people are much more likely to pick the first option, showing that they are risk-averse
When are people likely to be more risk-seeking?
People are more likely to be risk seeing when presented with a definite loss versus a risky loss.
- e.g. Definitely lose $50, or enter a 50/50 chance to lose $100 or nothing?
—According to utility theory, these two options have the same utility. However, people are much more likely to take the 50/50 chance, showing that they are risk-seeking
prospect theory
that the impact of a gain or loss depends on its relative value, and that the utility of an outcome depends on subjective value and subjective probability
- Expected value does not explain the psychological value between a gain and loss
—People hate losing more than they like winning
—”Losses loom larger than gains”
What does the phrase “losses loom larger than gains” refer to?
This phrase refers to the fact that people dislike losing a lot more than they like winning. For example, people would take larger risks to avoid losing than they would to potentially win more.
- As the relative value of a gain or loss increases, the psychological impact of a gain or loss diminishes
—e.g. The difference between $0 and $100 seems like a lot, but the difference between $500 and $600 feels like less. This is because it is more psychologically impactful to gain or lose $100 from $0, versus gaining or losing $100 when you already have $500.
How do framing effects challenge utility theory?
Framing effects demonstrate how people make decisions based on subjective thinking
—Framing effects tend to interfere with people’s ability to make logical decisions
—If people were always rationally, or thinking in terms of expected value, then they would be affected by framing effects
framing effects for gains vs. losses
- people are risk-averse for gains
- people are risk-seeking for losses
—People prefer a sure gain over a risky gain (e.g. gaining $10 for sure, versus either gaining $20 or nothing) but people prefer a risky loss over a sure loss (e.g. losing $20 or nothing, versus losing $10 for sure)
—e.g. “Kenjaku will launch an attack on Shibuya, and it is estimated that some 1000 people will die unless other sorcerers intervene. Gojo has two choices. He may (a) evacuate civilians before the incident, thus surely saving 800 people. However, he will need to conserve his energy and will not be able to save anyone else during the actual attack. Or, (b) defeat Kenjaku with Hollow Purple in the beginning of the attack, thus killing off 200 innocent people in the process but allowing everyone else to live. Which option should he pick?”
—People are more likely to pick the first option because it is framed in “saving 800 people,” while the second option is framed in terms of “killing off 200 innocent people.” However, both options end in the same result: 200 people die.
framing effects for permission vs. prohibition
people may focus on different aspects of an option depending on whether they’re looking to grant permission for it or prohibit it—whether they’re looking to accept it or deny it
- e.g.
—Parent A: avg income, avg. health, avg. working hours, good relationship with child, stable social life
—Parent B: above avg. income, minor health issues, work-related travel, very close relationship with child, active social life
—When awarding a parent with custody, 64% of people chose Parent B.
—When denying a parent custody, 55% of people chose Parent B (still a small majority!)
—People will focus on positive or negative aspects of a problem depending on whether they have to decide to reject or accept it
certainty effect
Certainty has a stronger effect on people’s decisions than “less uncertainty”
- e.g. “You are choosing between two new vaccines. Vaccine A is $3000, and has a 90% efficacy. How much more would you be willing to pay for Vaccine B which has a 93% efficacy?”
VERSUS
“You are choosing between two new vaccines. Vaccine C is $3000, and has a 97% efficacy. How much more would you be willing to pay for Vaccine D, which has a 100% efficacy?”
—People are typically willing to pay more for Vaccine D than Vaccine B, even if the difference between 90 and 93 is the same as between 97 and 100. This is because “certainty” is more valuable than “less uncertainty”
framing effects for personal vs. impersonal
people’s decisions may be influenced by how much personal involvement a choice involves
- e.g. The Trolley Problem
—Would you rather pull a level to switch the track of the trolley, thus killing one person tied up on the tracks but saving 4 people, or push someone to their death off of a bridge onto the trolley, thus stopping the trolley’s path and saving 4 people?
—People prefer the first option because it has less personal involvement, but option 1’s outcome is the same as option 2’s.
framing effects for frequencies vs. probabilities
people’s decisions can be influenced by whether choices are framed as frequencies or probabilities
- Would you rather go to a bakery where 1 out of every 20 pastries on the shelf is underbaked, or a bakery where 5% of the pastries on the shelf are underbaked?
—These two bakeries are the same; 1/20 pastries (frequency of underbakedness) is the same as 5% of all pastries (probability of underbakedness) at the bakery. Yet, 5% seems like a smaller number than 1/20.
What does having more options do to people’s ability to make decisions?
Having more options typically makes people take longer to make decisions, makes people more likely to “wait” for other options, or makes people choose the option that requires the least “decision making”
- e.g. You’re doctor who has to decide between recommending drug A or getting surgery.
—You are more likely to opt for recommending drug A, since it seems less extreme than surgery.
- e.g. #2 You’re doctor who has to decide between recommending drug A, drug B, or getting surgery.
—You are more likely to opt for recommending surgery, since it is difficult to justify one drug over another
