expected value (EV)
- average outcome if a scenario is repeated many times
- calculated using probabilities and values of possible outcomes
- Ex. 75% chance of winning $200, 25% chance of winning $0.
- EV= (.75 x 200) + (.25 x 0) = $150
rational choice
in order to maximize outcome, choose greater EV
- A: win $125
- B: 25% chance of winning $400, 75% chance of winning $0.
- EV= (.25 x 400) + (.75 x 0) = $100
- choose A
advantages with using expected outcome
advantages
- clear prescription for correct choice
- leads people, on average, to maximize monetary gains given what they know about the world
- keeps people’s decisions internally consistent
problems with using expected value
- difficult to apply for non-monetary decisions
- doesnt explain actual choices by actual people
prospect theory
- daniel kahneman
- people do not make decisions based on EV, probabilities, and absolute outcomes
- people make decisions based on subjective utility, decision weights, and relative outcomes
subjective utility
- people transfer objective value into subjective utility (usefulness of outcome)
- diminishing marginal utility : subjective utility increased more slowly than objective value, esp. at large values
- individual differences
loss aversion
- losses loom larger than gains
- losing $20 feels worse than winning $20 feels good
decision weight
- people transform objective probability into subjective decision weights
- small probabilities (>0%) are overweighted: 1% feels like 0, 51% feels like 50
- large probability (<100%) are underweighted : 99% feels a lot less than 100%, 50% feels same as 51%
framing effect
- people make decisions based on gains and losses relative to a point of reference, not based on absolute outcomes
- changing the way a question is asked to create a different point of reference leads to different valuations and thus different choices
people make decisions based on individual: (3)
- subjective utilities : diminishing marginal utility & loss aversion
- decision weights :
underweight large probabilities & overweight small probabilities - relative outcomes :
reference dependence, gain & loss framing
dopamine
- schultz, dayan & montague
- single until recordings from monkey’s midbrain dopamine neurons in ventral segmental area
learning driven by rewards
- activity of midbrain dopamine neurons is related to reward
- but dopamine neurons do more than simply report occurrence of reward
- they code deviations from predictions about time and magnitude of reward
prediction error
PE = actual reward - expected reward
>0, better than expected
= 0, as expected
< 0, worse than expected
dopamine pathways in human brain
-midbrain dopamine neurons project to basal ganglia, prefrontal cortex, and other areas
Obital frontal patients : emotion and decision making
- bechara 1994 : compared control participants and patients with damage to orbital frontal cortex
- patients : perform normal on IQ tests, normally on tests of cognitive control, seem to make poor decisions in life
Iowa gambling task & OFC patients
- goal to win max amount of money
- compared control to OFC patients, used 100 deck choices
- asked what they thought about decks, recorded choices, measured skin conductance at time of decision making (e.g. sweaty palms)
Iowa gambling task
- conceptual knowledge alone did not lead to good decision making
- predictions of expected emotions necessary for good decision making
- expected emotions receded conceptual understanding
- OFC patients could not generate expected emotiong
- patients overemphasized immediate reward over long-term outcomes
