Heuristics and Biases Flashcards
Background
Heuristics— mentsl shortcuts, rough and ready procedure for making a decision
—use them without comparison of other options
—might not have time/resources to make other choices— small decision
—dont guarantee we will come to optimal or best result- may or may not be effective
Herbert Simon- consequence of bounded rationality— consequence of bounded rationality- human decision makers cant be fully rational
Satisficing— first heuristic— choosing good enough decision
History
1970s— Tversky & Kahneman (T&K)
Gerd Gigerenzer— german— according to him— heuristics can lead to excellent decisions— fast and frugal— but also very efficient
Representativeness heuristic
People us this when judging probability that event a originates from process b
—use resemblance as a proxy for probability judgement
—sometimes can lead to wrong choice— serious errors— similarity— neglecting important factors when basing judgements on similarity or representativeness
Base Rate Neglect
(A)
—K/T showed people a short personality description — then asked ppl to asses probability that it belonged to an engineer or lawyer
—Dick, 30, successful in field, married no children, high ability/motivation, liked by collegues— issue description— doesnt convey ny relevant info— absence of any other cue— 50/50
p = .5 regardless of base rate
Sample 1 group told— sample 70 lawyers, 30 engineers
other groop— 30 law, 70 eng
Result— essentially same prob— no effects on judgement
Sample size neglect
Representativeness (B)
K&T— men drawn at random from population of 1000, 100 and 10 men is taller than 6ft tall
Probability of man being 6 feet tall was the same regardless of sample group
—this must be wrong— bigger the sample, more likeliness of 6ft taller
Coin toss— toss it 5 times or 50–what is likelihood of getting heads or tails
Regression Fallacy
Representativeness (C)
K&T noticed that flying instructors if they gave trainee a praise for a exceptionally smooth landing was usually followed by rougher landing on next attempt
—vice versa
-any random deviation from the mean is likely to be followed by one closer to the mean
think of police— cracks down on sudden increase in specific area— then crime goes down later on— can be effect of police but also phenomemon related to regression of the mean
Availability (A)
People often judge the probbility of an event by the ease with which they can think instances of it
—can cause bias: availability is affected by external factors— might not be relevant for current decision making
first, retrievability
—K/T saw list of famous people in public life— ppts asked to judge whether the list contained more men >women 1973
—set up so some lists—men more famous than women vice versa— ppts asked more women/men— not famous rank
—ppts using persons fame to remember how many people—gender with more famous names— more numerous
Availabilty (B)
Effectiveness of search set
K/T asked people— suppose we sample a word of three letters or more at random from english text— more likely word to start with K or K as a third letter
—found its easier for memory to list all word beggining with K
—using retrievability to make a judgement on frequency
third, imaginability bias (consequence)
—TK asked ppl to judge how many diff grps can be formed from 10 ppl- 2,4,6,8,
—easier to imagine groups of 2 than of 8– estimates decrease with size—
—med estimate 20– grps of 8– 70 for grps of 2
Availability (C)
illusionary correlation (4th)
—introduced by Chapman & Chapman— employed clinicians and students— given them info ab patients, descritions, diagnostic statements and drawings of person made by patient
—later judged the frequency with which each diagnosis (paranoia) — accompanied by various features— peculiar eyes
—ppts greatly overestimated the co-occurance of peculiar eyes etc— even when they were actually negatively correlated
—interpreted TK that availability— ease of recalling examples that fit prior beliefs, cause bias— completely disregarding when separate
