Decision making Flashcards
What is certain vs. uncertain decisions?
Certain decisions are choices where the outcomes are known and predictable.
Uncertain decisions involve choices where the outcomes are unknown or probabilistic. (For example, gambling)
What are prescriptive vs descriptive norms?
Prescriptive norms tell us what we should do, based on social rules or expectations (e.g., “Say thank you after receiving a gift”).
Descriptive norms show what people actually do, based on observed behavior (e.g., “Most people leave tips at restaurants”)
What is expected value theory?
A prescriptive model, where people choose the option with the highest expected value. They make rational decisions to maximize long-term gains.
“How people should behave if they are maximizing their expected value”
“Subjective value = Objective value” means people value outcomes based purely on their actual worth, without personal bias (100$ is the same for everybody)
What is expected utility theory?
A descriptive model, where people choose the option with the highest expected utility (personal satisfaction), not just objective value.
Subjective value = f(objective value): Personal value depends on the objective value but is shaped by individual preferences.
Subjective utility (how useful is this outcome to me) - decisions are based on subjective representation of value (utility) rather than the true objective
Risk preference: Choices reflect whether someone is risk-averse, risk-neutral, or risk-seeking
What is Prospect theory?
The best one:
Focuses on how people evaluate gains and losses relative to a reference point, with a strong emphasis on loss aversion (losses hurt more than gains feel good)
Changing our proposal about how value are related to utility making the curve steeper for losses - loss aversion.
Rather than focusing on the absolute amount of money we have, now the curve can shift around according to a certain reference point (so the curve crosses the x-axis different places depending on the amount of wealth you have
Through parameters like:
The frame = The way options are presented affects decisions (e.g., “90% success” vs. “10% failure”)
Framing effect = changing the reference point
Endowment = People value things more once they own them (people are willing to buy a mug for 3$ but sell it for 4$ - the positive utility of gaining a mug is equal to the negative utility of loosing a mug)
Describe utility vs value
Utility is about personal satisfaction or benefit
Value is about how much something is worth, either objectively (based on price or function) or subjectively (based on personal preferences)
What are the different risk preferences?
Risk-averse: Prefers safe choices, avoiding risk.
Risk-neutral: Focuses only on the outcome, not the risk.
Risk-seeking: Prefers risky choices for the chance of bigger rewards.
What is loss aversion?
The idea that losses feel worse than gains feel good. People are more upset by losing something than they are happy about gaining the same amount
What is a value function?
Assigns a value to different choices based on their expected outcomes or rewards.
It typically shows that losses feel stronger than equivalent gains (loss aversion) and that the value of gains or losses decreases as they increase (the bigger the gain or loss, the less it emotionally impacts you)
What is a utility function?
Shows how much satisfaction a person gets from different outcomes. It helps guide decision-making by comparing the value (utility) of each option
What are framing effects?
Framing effects happen when the way information is presented influences decisions
Win frame = when a situation or decision is presented by emphasizing the potential gains or positive outcomes (“you have a 90% chance of success” )
Loss frame = emphasizes the potential losses or negative outcomes of a decision (“you have a 10% chance of failure”)
Who was Phineas Gage?
A man who survived a severe brain injury when an iron rod pierced his skull. Despite surviving, his personality changed dramatically, providing insight into the role of the frontal lobe in personality and behavior
Describe the brain areas involved in decision making
Prefrontal Cortex: Handles planning, reasoning, and evaluating long-term consequences
Ventromedial prefrontal cortex: Tracks probability of reward, helps us think carefully about rewards (involved in gambling behaviour)
Orbitofrontal Cortex: Assesses the value of outcomes and helps make decisions by evaluating rewards and punishments
Basal ganglia: tracks subjective utility - dopamine neurons in the basal ganglia react quickly and automatically to rewards
What is Bayes’s Theorem/rule?
A formula in probability:
A prescriptive model, that helps update the probability of a hypothesis by combining prior knowledge with new evidence to calculate the updated (posterior) probability.
It helps make rational decisions by updating beliefs with new information, where you take the ratio into consideration (farmers vs librarians ex.)
Mark used examples like the rare desease and spam filter
What is the rare deasease example?
Imagine a rare disease that affects 1 in 1,000 people. A test for the disease is 99% accurate (it correctly identifies people with the disease 99% of the time). If you test positive, Bayes’s Theorem helps you calculate the actual probability that you have the disease, considering both the rarity of the disease and the test accuracy.
Even with a positive test, because the disease is so rare, the chance you actually have it is still lower than 99%
What is the spam filter example?
Bayes’s Theorem is used to calculate the probability that an email is spam based on certain features, like specific words or phrases. The filter combines prior knowledge (how often certain words appear in spam) with new data (words in the incoming email) to update the likelihood that the email is spam. This helps the filter make a more accurate decision about whether to mark the email as spam or not
What is prior probability?
The initial likelihood of a hypothesis before considering new evidence
What is conditional probability?
The probability that a particular type of evidence is true if a particular hypothesis is true
For example, if you know an email is spam, the probability of it containing certain words is the conditional probability.
