decision making

Question 1

DRM paradigm

Answer 1

using schemas to infer things

source misattribution or source memory

Question 2

TAP

Answer 2

transfer appropriate processing
* levels of processing leading to better memory
* refute it because the way you process it matters
* rhyming at encoding and at test results in better memory

Question 3

what distinguishes semantic from episodic memory

Answer 3

time and place
* connected autobiographical memory (combination of the two)
* both explicit
* semantic knowledge comes from episodic memory

Question 4

decision making

Answer 4

what is it: choosing options on the basis of the information given which is often incomplete

why would information be incomplete
* because it is not available
* because there isn’t time to consider it all
* because short term memory can’t hold it all

what information do people use and what do they ignore

psychology and economics connect
* never a csae where you have all the information and the information is incomplete so you have to make judgements on which info to prioritize and which to ignore

Question 5

economic models

Answer 5

economics and businesses are intensely interested in how people make choices

economics traditionally relied on prescriptive models:
* expected value
* expected utility

a prescriptive model tells us how we “ought” to make decisions

a descriptive model says what people actually do

** this is the ideal and rational way to make decision
** how people ought to make decisions
* people don’t always make decisions in the most optimal way

Question 6

expected value example

Answer 6

which one would you buy

choice of 1 or 2 lottery ticket
* (A) Probability is 1.0 (a guarantee) of winning $1
* (B) probability is .00000014 of winning $3million

according to expected value:
* (A) is worth $1
* (B) is worth 42 cents

if everyone used expected value, there would be no lotteries
— so…this model is almost certainly wrong

Question 7

expected value

Answer 7

• the expected value model is normative: defines the ideal performance under ideal circumstances
• expected value: probability of outcome times absolute monetary value of outcome (probability) x ($)
• assume an expected value model
  – for each choice people weight the value of the expected outcomes
  – then chose the outcome with highest expected value

expect the person to choose the highest expected value
* but doesn’t happen when you test it

Question 8

expected utility

Answer 8

with expected utility, people chose an alternative with the highest utiilty or benefit

• for each choice, people weight the value (+ or -) or the expected outcome
• like expected value, but not necessarily dealing with money (probability) x (utility)

what kind of model is it?
* normative model

– allows to look at decision making outside of monetary options
– more broad options to research
– normative bc capturing how most people would decide on a couple of options

Question 9

what do people really do when making decisions?

Answer 9

1. we tend to be loss aversive
   * loses are weighted higher than gains
2. the starting point (“anchors”) influence decisions
3. the framing of a problem is critical (options given will affect the final choice made)
4. people avoid risks (tend not to choose risky options)
5. we use heuristics (short cuts)

** moving away from monetary to utility allowed to capture more
** but people didn’t follow monetary or utility

depending how you frame the problem results in difference choices
** shortcuts that get you to an answer quickly

Question 10

loss aversion

Answer 10

Kahneman, Thaler (1990) divided class into two groups
* sellers are given a mug to keep and asked how much they are willing to sell it for
* choosers are asked how much money they would find as attractve as the mug

same situation for both but perspectives differ
– sellers lose their mug and gain a mug

Question 11

loss aversion results

Answer 11

median prices
* sellers $7
* choosers $3

we have tendency to place higher value on things that are ours

Tversky & Shafir (1992) found that many people reject a 50-50 bet in which they can win $200 but lose $100 – economic theories can’t explain this

Question 12

Status Quo

Answer 12

loss aversion tends to maintain the status quo

samuelson (1994) told people they had inherited a large amount of money as
– blue-chip shares, risky shares, T-bills, or bonds

each group presented with the same amount of money in one of these four forms

they can take the money anyway

Loss aversion is to maintain status quo – strong tendency to do so

Question 13

example of status quo

Answer 13

new jersey and pennsylvannia have both introduced a limited right to sue, which will lower your car insurance

In new jersey motorists pay for the right to sue (only 20% of new kersey drivers acquire it)

in Pennsylvania, the full right to sue is the default (75% of Pennsylvania drivers retain right)

people are less likely to opt into something than they are to opt out

Question 14

anchoring

Answer 14

when making estimates people start with an initial estimate then adjust it

the position you start from, influences adjustment

related to loss aversion and status quo
* if you’re going to move from your original position, you don’t move very much

Question 15

anchoring example

Answer 15

subjects told in 2 seconds estimate product of multiplications

A: 1x2x3x4x5x6x7x8
B: 8x7x6x5x4x3x2x1

people who start off with low numbers make an estimate that is way smaller than those who start with high numbers

another anchoring example
– study of real estate agents shown the same house and asked to make an appraisal
– first group told the asking price was way less than second group – so second group offered more than the first group

if you sell something at a discounted price, people will be more willing to think it’s a fair price

Question 16

designing a restaurant’s menu

Answer 16

expensive items on a restaurant menu help sales even if no one orders them (steer customers to pricier (not not as expensive) items

menu decoy items – some restaurants put expensive items at the top of the menu and make the others seem less expensive
– research shows most people order neither the most nor the lest expensive dish, so strategically-placed decoys can boost sales of other items

Question 17

framing effects

Answer 17

the way a problem is framed can change the way the options are evaulated

imagine that Canada is preparing for the outbreak of a new disease, which is expected to kill over 600 people. two alternative programs have been proposed. assume that the exact scientific estimates of the consequences are as follows:
— framed pictures
— framed paintings
— framed art

Question 18

framing effects first and second set of options

Answer 18

option set #1
* program A: 200 people will be saved
* program B: 1/3rd probability that 600 people will be saved, 2/3rds probability that no people will be saved

option set #2
* program C: 400 people will die
* program D: 1/3rd probability that no one will die, but 2/3rd probability that 600 will die

more people choose D than C because they are framed differently
– if you frame it as people surviving vs people dying the results are different
– people chose the one that will save the most people

people don’t use utility to make decisions

Question 19

risk aversion

Answer 19

• depending on how is presented, different options seem more risky
• people tend to avoid the risky choices (choices B and C seem more risky)

experienced doctors vary in choice of alternative treatment for lung cancer (radiation treatment or surgery) depending on whether mortality or survival rates are given

Question 20

heuristics

Answer 20

strategies that can be applied easily to a wide variety of situations and that often lead to reasonable decisions

but not guaranteed to work: they will often lead is to the correct solution but not always

** to come up with a solution faster even if its not ideal

Question 21

availability heuristic

Answer 21

judgments based on ease with which relevant instances can be retrieved from memory

example:
1. estimate in 2 seconds how many flowers you can name in ten minutes
2. estimate in 2 seconds how many Russian novelists you can name in 10 mins

likelihood of event is evaluated by how easy it is to come up with examples

making decisions based on how available that information is to you

Question 22

availability heuristics example

Answer 22

asked estimated responsibility for household activities for 37 married couples

estimated contribution of household activities for themselves
* estimate more for their own contributions
* spouse overestimate their contributions

Question 23

representative heuristic

Answer 23

look for things that make semantic sense even if the probabilities don’t match

if something or someone appears to fit a category, you will judge them to be part of that category
– racial profiling

“Linda is single, outspoken, and very bright. She majored in philosophy. In college she was involved in several social issues including the environment, peach campaign and the anti-nuclear campaign.”

which is more likely:
A. Linda is a bank teller
B. Linda is a bank teller active in the feminist movement

people vote B more likely even though probabilistically, A is more likely

Question 24

implicit bias

Answer 24

discriminatory biases based on implicit attitudes or implicit stereotypes

implicitly biases are especially intriguing and problematic, because they can prodice behavior that diverges from a person’s avowed or endorsed beliefs or principles

based on assumptions of another person based on the category you put them in

bias comes out in people’s behavior even if they say they don’t have bias

automatically activated evaluations outside of a person’s awareness (unconscious)
* formed slowly through experience
* resistant to extinction

Question 25

stereotypes

Answer 25

• cognitive structure that contains perceiver’s knowledge, belief and expectations about a social group
• A stereotype assumes that all members of a group share some common feature

stereotypes, biases, and shortcuts all show that we always are categorizing stuff and people

Question 26

racism and sexism

Answer 26

modern racism: prejudice directed at other racial groups that exists alongside an explicit rejection of racist beliefs

benevolent racism and sexism
* bias in unstated and hidden
* researchers also refer to hidden bias as cognitive, automatic, or implicit bias
* it is nonetheless powerful and pervasive

Question 27

gender stereotypes

Answer 27

• women are kind, communal, care about others, pay attention to emotions
• men are agentic, task-oriented, have leadership qualities

backlash when one gender takes on stereotypical characteristics of the other gender
* women taking leadership roles
* if women are not warm, they are selfish and cold

stereotypes do not fit with how most humans are

Question 28

gender bias

Answer 28

psychological “fit” between gender stereotypes and role stereotypes (women and leadership)

gender prejudice emerges from the clash of gender stereotypes and work role expectations: role incongruity or lack of fit

leadership roles are equated with masculinity. female leaders are evaluated as less leader-like than their male counterparts. Backlash against female leaders

Question 29

how are stereotypes and biases formed

Answer 29

social learning process
* parents
* significant others
* media
* systems of education, politics

categorization processes
* cognitive efficiency
* understanding and prediction
* enhance social identity

Question 30

disrupting implicit bias

Answer 30

Jennifer Eberhardt (Stanford)

• slow down – thinking through the reason why a decision is being made
• creating friction – forcing people to provide justification for the decision

your conditioning comes first, your thinking comes second

Question 31

Gambler’s Fallacy

Answer 31

If I flip a coin 8 times, which is more likely

Another Example: A die with 4 grey sides and 2 red sides is rolled several times

you will be paid $25 if correctly choose the sequence
* RGRRR or GrGRRR

why is it called the Gambler’s Fallacy
* think of a roulette table: the last sequence was Red Red Red Red
* so which is more likely to come up next?

Question 32

biases vs heuristics

Answer 32

Heuristics: simple cognitive rules that are easy to apply and that usually yield acceptable decisions but can lead to errors (shortcuts)

biases: systematic errors that result from using heuristics in decision making

Question 33

hindsight bias

Answer 33

“I knew it all along”
“it was inevitable”

once we have learned the facts of a case, we tend to believe we knew it all along

can make use think (erroneously) that we didn’t need to do the research

Question 34

Hinsight Example

Answer 34

experimental group
* given the asnwers in advance
* then asked how confident they were that they would have gotten it right had they not been given the answer

experimental group showed much greater confidence than Control group
* also happens for football games, juried trials, political races, and political hearings

Question 35

overconfidence bias

Answer 35

confidence in decisions continues to climb as more information is obtained

this bias is most extreme in tasks of higher difficulty
* estimating our potential productivity
* I can do the assigned paper in 3 hours no problem

values for confidence are systematically higher than actual accuracy

Question 36

saliency bias

Answer 36

• more vivid or salient events are remembered better, thus biasing our judgment of how often they occur
• which heuristic would this be associated with

Question 37

familiarity bias

Answer 37

judging an event as more frequent or important just because it is more familiar in memory

Question 38

familiarity can pay off

Answer 38

Gigerenzer: had to group pick selection for stock portfolio
* experts (highly experienced stock brokers)
* novices (university students

results: stock portfolio picked by novices based on familiarity outperformed stocks chosen by experts

Question 39

other factors that affect decisions

Answer 39

Illusory correlations: a perceived correlation, or relationship between 2 variables that does not really exist

regression to the mean
* if an outcome is extreme, the next outcome will be closer to the population mean

Question 40

illusory correlations

Answer 40

in any random sequence, there will be patterns that appear to be meaningful

• illusory correlations occur because the brain is constructed to search out meaningful patterns, and make sense of experiences

so sometimes it does this even when experience is not meaningful (Example: lucky socks)

tendency to see meaningful connections when there are none

Question 41

Illusory correlations – examples

Answer 41

the hot hand myth

belief that a professional basketball player is more likely to make a shot after they have just made one than after they have missed one

Question 42

regression to the mean

Answer 42

suppose an educator tests a class to see who could benefit from remedial instruction

• the selected kids improved on post-test
• remedial instruction confounded with regression to the mean

other examples: one hit wonders, sophomore slump

As sample size increases, the sample is more likely to be representative of population
* another example: at the end of April, someone is almost always batting, 400 in the major leagues
* but no one has finished season about .400 since 1941

so, as sample gets larger, players tend towards career averages
* predicts that free agents often disappoint

Question 43

regression to the mean cont’d

Answer 43

makes it look like punishment works

example: airforce, instructors believed that punishment was better than praise

reason: when student did really well early on, they praised them,, but then they often started doing worse
* when student started out doing badly, they yelled at them, and they often got better

• predicted by regression to the mean

Question 44

gambler’s fallacy

Answer 44

each flip is random so unlikely to get flip you want

judgements are not true reflections of the probability of the situation
* coin flips are random, so expect sequence to reflect randomness

errors like this result from the shortcuts you take (Heuristics)

representativeness heuristics bc you are judging the probability of an event by assessing it to how similar it was to previous experiences

Question 45

regression to the mean class explanation

Answer 45

happens just because of how statistics work but we attribute different meaning to it

the more samples you add when measuring, the more likely you’ll have a central tendency of what is is

imposes itself on how people make judgements

selectings kids based on their performance on one test, getting help, and then post performance test
** could be that the help actually helped but could also just be a regression to the mean, where they were just worse in the first test than the second

Example
– performing high beginning of season and then regressing back to the mean mid season