Decision making Flashcards
What would be ideal way to make decisions?
Expected utility theory–designed to make idea decisions; about Maximizing value/utility
Von neumann and morgenstern
Expected value= odds of gain * value of gain
Odds between zero (0%) and 1(100%)
Problem with expected utility theory is ppl dont really behave that way
Humans are not computers (or casinos wouldnt work)
decision making (basic concept)
Human mental life can be trisected into past (memory), present (perception/continuous flow of info from environment) and future
Humans are just as good as projecting into future as into past
We can cast mind forward to most plausible future and/or hypothetical situations
Organisms can respond (basic hardware), predict (pavlovian software) and operate (skinnerian software)
Prospecting–
uniquely human form of intelligence that enables us to think about far future events
So how do we actually make decisions?
Judgment under uncertainty: bugs come from mental heuristics (simple strategies or mental processes that humans, animals, organizations and machines use to quickly form judgments, make decisions, and find solutions to complex problems.)
We do this estimation all the time/approximation (8797; we do 87100 as a shortcut and end up with the wrong answer)
Two bugs in our system: errors of odds and errors of valuation
Errors of odds
Likelihood of success diff from actual likelihood
Sample size neglect, gamblers fallacy, conjunction fallacy, availability bias, planning fallacy
Sample size neglect:
you would think that sample sizes produce same results
gamblers fallacy,
predict based on last thing
Belief that likelihood of chance event influenced by nature of events that preceded it
Conjunction fallacy-
when u add second possibility to first ur always decreasing
The `Conjunction Fallacy’ is a fallacy or error in decision making where people judge that a conjunction of two possible events is more likely than one or both of the conjuncts.
(if linda is SJW we assume that she’s feminist so when ppl are asked is linda more likely to be bank teller or bank teller and feminist ppl say bt and feminist)
Availability bias–
overestimate frequency of things that are more easily available
Planning fallacy–
world always delivers a little more negative/people are bad at estimating delays, price–easy to imagine sitting down and putting in a good day of work but harder to conjure all of the distractions–end up thinking they wont get in the way
Errors of valuation
thinking value is different
Presentism, comparing to past, prospect theory, temporal discounting
Presentism
tendency to overvalue things in the present–willing to pay more for a dinner tomorrow if ur hungry–bad idea to grocery shop while ur hungry
Comparing to past–
two job offers for 3 years–(even tho salary decreases each year u get more, but people dont like to feel worse off so they choose the increase)
Prospect theory–
kahneman and tversky looked over examples of errors and said ppl arent using expected utility so can we come up w another
Subjective value (how valuable is smthn to you) and objective value (actual cost)
If we were super rational there would be perfect correlation
Actual human cognition is a curve line–non linear relationship between objective and subjective and diff between 10 and 20 dollars subjectively seems like a bigger diff between 1k and 1,010 dollars even tho its same
Curves asymptote–levels off
Steeper for losses than for gains–losses loom larger than gains–losing ten feels worse than gaining ten dollars feels good
Endowment effect: objective value of mug $5 peoples subjective value (to sell back) is $7 but buyers bid was $3
Mug worth more for someone selling than buying; loss looms larger than gain
Not a rational system but its how we do things
Temporal discounting
Gains at diff times
more>less, sooner>later
$10 now, or $12 in two weeks? Ppl choose ten dollars now–12 dollars become less valuable over time
We discount over time
Happy to commit to waiting a week a year from now but won’t commit for now
We can’t imagine things far off
