Decision-Making

Created by Axel Burch

decision-making
>reasoning about choices >each choice has costs and benefits

1/42

TermDefinition
decision-making
>reasoning about choices >each choice has costs and benefits
costs
take us away from our goal
benefits
take us towards our goal
value/utility
weigh up benefits against costs to maximise value
normative theory
>how decisions ought to be made >assign different values to things >imagine decision outcomes and their consequences >judge usefulness of things
SEU calculation
(Probability1xUtility1)+(robability2xUtility2)
subjective expected utility theory
>normative, mathematically-oriented theory of decision making >weigh up attributes according to probability >choose option based on attributes best suited to needs, given constraints >aims to maixmise utility
descriptive invariance in SEU
description of decision task options should not influence option chosen
evidence of descriptive invariance
(Tversky and Kahneman) >scenarios identical in 'expected utility' but responses differed depending on 'gain' or 'loss' emphasis >choices affected by framing effect
framing effect in prospect theory
>risk-seeking decisions more likely with negative 'frame' >risk-averse decision more likely with positive 'frame' >violates invariance principle
evidence for framing effect
>Asian disease problem >health promotion problems
prospect theory
>descriptive theory of decision-making >explains how decisions are made, and how decision deviate from normative standards >combines probability and utility
cognitive heuristics
>reflexive mental shortcuts that simplify complex cognition >increase speed and efficiency of thinking >reduce cognitive load >can lead to bias/error as not all relevant information considered
heuristics
mental strategies used under certain conditions due to cognitive limitations, external constraints, or motivational constraints
original approach
(Kahneman and Tversky) >challenged dominance of normative models of thinking >thinking under conditions of uncertainty not due to extensive information, but use of simplifying mental strategies >cognitive model of bias in judgement and decision-making
types of cognitive heuristics
>availability >representativeness >anchoring and adjustment
availability heuristic
judging something based on ease of recall or mental stimulation of consistent evidence
representativeness heuristic
judgements based in similarity between seemingly related things
anchoring and adjustment heuristic
judgements based on numerical information
evidence of the availability heuristic
>memory retrieval/mental stimulation of event-consistent information leads to bias >e.g. reading news article about house fire lead to overestimation of occurrence chance
evidence of the represetativeness heuristic
>violate normative principles when reasoning probabilistically >'taxi cab' study- overestimation of probability that blue or green taxi involved in incident >links to conjunction fallacy
conjunction fallacy
>thinking error occurring with probabilities >2+ non-perfectly correlated items judged as being more likely to occur together than separately >violates 'extension rule'
evidence for conjunction fallacy
>'Linda problem' study >given information about characteristics, job, and interests >judge likelihood of information about 'Linda' >found 2 pieces more likely to occur together than any piece separately with their judgments
extension rule
the probability of 2+ items occurring together can't be more likely than each of those events occurring separately
evidence of the anchoring and adjustment heuristic
>wheel of fortune study >Judge percentage of African nations in UN affected by ‘random’ number from ‘wheel of fortune’ beforehand >judgments higher with high anchor value, so biased in direction of anchor due to insufficient adjustment
biases
>overconfidence >over-optimism >explained by neglect of base-rate data >linked with case-specific reasoning
planning fallacy bias
>underestimation in laboratory and real-world tasks due to anchoring and adjustment heuristic >reduced by prior task experience >previous task performance as anchor for predictions
gambler's fallacy bias
>tendency to look for order or sequences >operational when judging repeated stimuli >misconception of 'randomness'/'chance >due to representativeness heuristic
rationality of heuristics
>'adaptive' critique >heuristics part of rational thought >ecologically-rational and serves needs to function >underpinned by bounded rationality theory and satisficing
adaptive approach
>heuristics and biases approach too focused in probabilistic thinking >frequencies more evident in real world >hard-wired to think if things in terms of frequency
frequencies vs. probabilities
>Gigerenzer et al.- cognitive biases disappear when probabilistic information presented in frequency format >Griffen and Buehler- no effect of information presentation on base-rate data neglect bias
cognitive toolbox
contents used to make effective judgement and decisions in real-world environments
fast and frugal heuristics
>dealing with uncertain situations >rarely lead to bias >serve everyday needs and goal
Gigerenzer et al.'s approach
>JDM not always error-prone in uncertain conditions >rational judgments and decisions often made in uncertain conditions >decisions made on basis of quick mental simulation of outcomes (satisficing)
recognition heuristic
>basis for judgments and decisions when knowledge is limited >'recognition' of relationship between cues in environment
'recognition'
ecologically-rational JDM method when knowledge is limited and ignorant of topic/task
recognition heuristic evidence
(Ayton and Onkal) >no effect of prior knowledge on accuracy of judgments >recognition of items useful in judgements
'take-the-best' ecological heuristic
using the best cue from the environment
'elimination-by-aspects' ecological heuristic
choosing by discriminating among attributes
revised approach
(Gilovich et al.) >reviewed 20yrs of heuristics and biases research >original approach too descriptive and didn't identify cognitive mechanisms
automatic vs. deliberate choice heuristics
>automatic (original) uses system 1 reasoning to save cognitive resources >deliberate (fast and frugal) uses system 2 reasoning to reduce cognitive load
affect heuristic
>identifying negative/positive stimuli quickly >feeling states guide thinking under conditions of uncertainty