PSYC 335 - Decision Making

Lecture 2 to 4

What is a decision?

Something with more than 1 possible response

Risky Decision

The outcomes of the possible responses are risky/uncertain

Possible decision parameters

Size of the outcome, probability of outcome occurring, time to outcome, effort to obtain

Economic definition of risk

Variance, +10/+90 is more risky than +40/+60 even if avg outcome is the same

Risk as hazard

Losing something is considered “risk”

Risk vs uncertainty in Knightian terms

Risk means you know the probability, uncertainty means you can only estimate or learn the probability

Utility

A common currency for choices, represents perceived value/perceived usefulness to an individual

Expected utilities

What is the expected return of utilities after making a decision

Bernoulli Effect

More wealth = less utility, less value the more wealth you get

Framing effect

How you frame a percentage can change how it is viewed

Tropical flu problem

Framing 1 version as saving people, one as fatalities

How does the tropical flu problem show the framing effect

If you frame as saving people (gains), people choose the less likely to choose the risky version, if you frame as preventing casualties (loss), people prefer the risky decision

Value Function

A function that plots utility vs gains/losses that everyone implicitly has in their head. Based on some amount of gains, you will have some amount of utility (i.e how much something is worth to you mentally)

What is an “ideal” value function

y=x

How does the value function relate to the framing effect?

If something is framed as a gain, the utility is lower than the utility from the same loss (i.e at x = 2 and x = -2, |y| is greater at x = -2) due to flattening

What does flattening of the value function imply

Since the gain flattens faster, more gains means a slower increase in utilities.

Prospect theory

Gains and losses are different parts of our value function, and the value function has a reference point, aka a point that decides if something will make you better off or worse off

Weighting function

Everyone has one, shows probability weight vs probability, essentially maps what people think the subjective probability of something is based on what the real probability is (someone might think 1% has a subjective probability of 5%)

How does our subjective probability generally compare to real probability

Generally distorted, we think low probabilities are more likely to happen, and high probabilities are less likely to happen

What was Lichtensteins study and what did it show

Asked people to estimate mortality rates of certain death causes

Cost benefit calculation

Making a decision based on analyzing the potential cost and the benefits before picking an option

Heuristic

A fast method to make decisions that are suited for the everyday world, and that are good enough usually.

Availability Heuristic

Mental shortcut that estimates frequency or probability by what comes to mind immediately

Representativeness Heuristic

Probability of an event is estimated from a similar situation

What kind of heuristic is the gamblers fallacy?

Representativeness, we estimate the probability to be 50/50 so we assume that after a streak one result is more likely than the other

Gamblers fallacy

Gamblers see streaks and ignore the independence of events and assume that a future result will be true based on past results

Conjunction fallacy

People don’t understand that the conjunction of two things (logical and) will be less likely, or equally likely than the individual events (P(a)+P(b) <= P(A))

Linda Problem

Subjects ranked different descriptions of Linda in order of most likely to least likely, shows the conjunction fallacy

Base rate neglect

People ignore base rates/basic frequencies in favour of similarity to reference category

Blackjack heuristics

People who increased bets after losses relied on representativeness, people who decreased rely on avaliability

What happened with Damasio’s patient, EVR?