TASK 5 - DECISION MAKING Flashcards
decision making
= process of selecting a choice from all the available options
stages of decision making
- theory of natural decision making
- setting goals: goals influence decisions; decisions are steps toward something you want to accomplish
- gathering information: know what the options, consequences are
- decision structuring: organise all information (listing options + criteria for deciding among them)
- making final choice
- evaluation
√ flexibility in phase order (if struggling, returning to previous phases)
cognitive illusions
= errors of cognition that come about for understandable reasons and that provide information relevant to understanding normal functioning –> systematic biases
- heuristics = rules of thumb; strategies that ignore part of the information with the goal of making decisions more quickly
cognitive illusions
- framing effect
- description frames the decision (e.g. 2 gas stations)
- “context effect” in decision making
- changing description can lead people to adopt different reference points
- loss aversion
cognitive illusions
- anchoring
= rely too much on an initial piece of information offered (= anchor) when making decisions
cognitive illusions
- sunk cost effects
= greater tendency to continue something once one has invested in it
- due to emotional investment, you are too invested in one ida and unwilling to quit
cognitive illusions
- illusory correlation
= seeing a relationship between variables even when there isn’t any
- typically have some prior associations in peoples’ minds
cognitive illusions
- hindsight bias
= tendency to exaggerate what could’ve been anticipated in foresight when looking back in hindsight
- once you know how a decision turned out, events leading up to outcome seem more inevitable than they really are
cognitive illusions
- confirmation bias
= only gather information consistent with prior hypothesis/information confirming previous idea
cognitive illusions
- overconfidence
= people’s impressions of own accuracy are inflated; arrogance
- fail to see need for help
cognitive illusions
- omission bias
= tendency to prefer inaction to action when engaged in risky decision making
- anticipated regret: greater when unwanted outcome caused by own actions rather than inaction
- status quo bias = decision avoidance; preference for maintaining the status quo (present state) rather than acting to change one’s decision
x action bias = don’t just sit there, do something
cognitive illusions/bounded rationality
- saitsficing
= strategy that aims for a satisfactory/adequate result, rather than the optimal solution
- “good enough” rather than expending effort/resources on finding best possible/optimal choice
1. satisficers = people content with making reasonably good decision; were happier and more optimistic, greater life satisfaction, less regret (not found in Chinese)
2. Maximisers = perfectionists - Simon
cognitive illusions
- fast-and-frugal heuristics
= heuristics involving rapid processing of relatively little information
- take the best, ignore the rest
x why do we have capacity of logical reasoning if we could always follow gut feelings
x more complex than suggested –> often not sufficient knowledge of validities
fast-and-frugal heuristics
- availability heuristic
= instances that are more easily thought of/accessible, stand out more in one’s mind
- own behaviours more prevalent to us than that of others
- failure to include base rate info in probability estimation
- availability by recall (number of instances recalled) vs. fluency mechanism (how easy it WOULD be to recall without actually retrieving from memory)
fast-and-frugal heuristics
- representativeness heuristic
= representatives/typical members of a category are encountered more often
- failure to include base rate info
- mistaken belief in the law of small numbers: people expect small sample to show same proportions as large sample
- gambler’s fallacy: mistaken belief that if something happens more frequently than normal during a given period, it will happen less frequently in the future
- conjunction fallacy: mistaken belief that the probability of a conjunction of 2 events (A & B) is greater than the probability of one of them
fast-and-frugal heuristics
- recognition heuristic
= using knowledge, that only one out of two objects is recognised, as basis for making a judgment; selecting object that is recognised in preference to the one not recognised (recognise cologne not Herne, decide cologne is bigger)
- exploits lack of knowledge
x more complicated than assumed –> consider why they recognise and object + decide then whether to use
base rate fallacy
= cognitive error whereby too little weight is placed on the base (original) rate of possibility
base rate information = the relative frequency of an event within a given population
utility models
= we try to maximise utility (subjective value we attach to an outcome)
- normative models = define ideal performance under ideal circumstances
- prescriptive models = show how we ought to make a decision (= consider that circumstances aren’t ideal and show how to best decide)
utility models
- expected utility theory
= trying to assess the subjective utility of certain objects
- normative model: ideal performance under ideal circumstances
- expected value (EV= ∑(p×v)): determine to choose the one with the higher value & see how much money you should spend
- expected utility (EU= ∑(p×u)) = (probability of a given outcome) x (utility of the outcome): captures ideas of happiness, pleasure & satisfaction that comes from achieving goals (= psychological rather than monetary value)
limitations of expected utility theory
x calculation of EU difficult: info about several aspects of the decision must be integrated (various factors and goals)
x only accounting for making final selection from a set of alternatives –> not making decisions with “status quo vs. make a change” option
x doesn’t describe processes by which people structure a decision (= gather info and lay out possibilities + parameters)
utility models
- multi-attribute theory (MAUT)
- ideal strategy –> bounded rationality (more realistic)
1. break a decision into independent dimensions
2. determine relative weights of each dimension (= subjective value)
3. list all alternatives
4. rank alternatives along dimensions
5. multiply ranking by weighting of each alternative to determine final value
6. choose alternative with highest value - normative model: if people follow MAUT they will maximise own utility in a way that is best for achieving all their goals
- dimensions must be chosen carefully
- person would be willing to choose an alternative that was not highest on the highest ranked dimension when the relative position on other dimensions compensates for that
limitations of multi-attribute utility theory
x elimination by aspects (descriptive model): when decision makers have too much info to deal with they don’t take all dimensions into account but rather select one factor, pick a threshold value and everything that exceeds that threshold value is tossed
x Homo economicus assumed: too rational
x procedure too complex
x set of relative dimension can not always be worked out; not clearly separate from each other
descriptive models
= simply detail what people actually do when they make decisions (= actual performance); what people ACTUALLY do
descriptive models
- bounded rationality
= people are as rational as their processing limitations permit
- produce reasonable/workable solutions to problems in spite of our limited processing ability (heuristics)
- unbounded rationality = if all the information you need, could choose best choice
- Simon
limitations for rational decision making
- cognitive constraints: limited capacities for information processing
- environmental constraints: not enough information about alternatives + consequences; limited time/resources
- -> use of heuristics
descriptive models
- image theory
- pre-choice elimination
= ask themselves whether a new goal/plan/ alternative is compatible with 3 images
1. value image: decision maker’s values, morals, principles
2. trajectory image: goals & aspiration for the future
3. strategic image: way in which decision maker plans to attain his/ her goals - incompatible with at least one of images drop out from further consideration
descriptive models
- recognition-primed decision making
= experts most likely rely on intuition, mental simulation, making analogies, recalling/creating stories when deciding; compare/recognise current situation to one encountered before
- expert decision making
- relevant when decisions must be made rapidly
- when situation familiar/typical: experts match situation to learned patterns of information stored in LTM (pattern recognition) –> retrieval of a single option –> mental stimulation –> if imagined outcome satisfactory that option determines actions
- when situation not familiar: diagnose situation further (story building, more data)
- Klein
descriptive models
- prospect theory
= decisions based on evaluation of potential value of losses and gains, using heuristics
- editing: identify reference point (subjective), generally representing current state
- evaluation: loss aversion
- Kahneman; graph
x partly explains value function
x does not include social/emotional factors
x framing depends on individual differences
x probability of rare events is underrated
descriptive models
- constructivist approach
= cultural representations are dynamic schemas rather than ever-present personality traits; switch between cultural frames in response to situational cues
- explains how people can be fluent in more than one culture without simply blending their biases
self-other decision making
= making choices for others involves less loss aversion than making choices for self
- making decisions for others gives us a feeling of power which makes us less sensitive to losses + seek more info when making decisions for others
- omission bias greater for self
self-construal priming on risk taking
= everybody has 2 co-existing selves in their memory: one independent + one interdependent
- cultural differences stem from likelihood of sampling one or the other (depending on social role & situation)
1. primed on an interdependent self: less likely to take social risks than those primed on independence
2. primed on interdependence: more likely to take financial risks - cushion hypothesis: social network acts as safety net; riends & family members offer a cushion that lessens effect of financial loss but magnify embarrassment of a social misstep
dual-process theory
- Kahneman
system 1: intuitive, automatic, effortless; heuristic-based
system 2: analytical, controlled, rule-based; more concscious
- serial processing
x assumes that people mostly use system 1
x oversimplification: people also rely on base-rate information in system 1/rapid judgments
x people can show strong biases even when using system 2 –> error-prone performance not only system 1
x serial processing is difficult to detect
endowment effect
= tendency for owners (potential sellers) to value objects more than potential buyers; simply owning an object can enhance its perceived worth
= self-object-association + self- enhancement bias
= enhancement of object’s perceived value
- smaller effect for Easterners (= faster in switching technologies etc.)
–> interdependent self: self-criticising
–> independent self: self-enhancing
natural frequency hypothesis
= better/more accurate judgment with natural frequencies than percentages
- people are naturally equipped to think about frequencies rather than probabilities
x benefits of frequency formats do not occur when using steps to avoid obvious version
x truly natural sampling may not be possible at all
x high demands on memory
rational-emotional model
= account for decision avoidance; decision Making based on rational factors
- omission + status quo bias: explained with regard to regret and fear (= fear can be reduced when someone decides not to make a decision at that time)
social functionalist approach
= takes into account social context of decision making
- people = intuitive politicians –> justify decisions to others
nudging
= steer people in a particular direction while still allowing them to go their own way (a subtle way of influencing behaviour); help them make good decisions without coercing them
- alter “choice architecture” (= background against which choices are made)
- default rules/ options: option a decision maker will obtain if he/she does nothing
- -> establish status quo/reference point for counting changes as losses or instead as gains
- -> depends on the costs of the decision + the cost of error
- -> often thought to be most effective nudge (fail when strong antecedent preferences or opposed with counternudges)
- Thaler
- uber tricks
nudging ineffective when…
x based on inaccurate understanding of behaviour
x info is too confusing/complex (unaffected by it)
x people show reactance: reject an official effort to steer just because it’s an official effort to steer
x they have short-term effect
x it influences desired conduct + produce compensating behaviour (rebound effect) –> nullifies overall effect
responses to failed nudging
- do nothing: failure might show that there is no problem at all
- nudge better/different when convinced that people’s choice is not promoting their welfare
- fortify/defend effects of nudge with a more aggressive approach (e.g. counter-counternudges or bans)
availability heuristic
- support theory
= event appears more/less likely depending on how it is described; distinguish between events and descriptions of those events
- more explicit event descriptions have greater subjective probability, BECAUSE
1. explicit description often draws attention to aspects of the event less obvious in non-explicit description
2. memory limitations prevent people from remembering all the relevant info if it is not supplied
x oversimplification
x explicit descriptions can reduce subjective probability if they lead individuals to focus on low-probability causes; if they are hard to understand
dual-process theory
- logical intuition model
= there is rapid, intuitive processing (system 1) of heuristic and intuitive logical processing (base rates) which is sometimes followed by deliberate system 2 processing
improvements to dual-process:
√ heuristic and base-rate information can both be rapidly accessed through intuitive processing
prospect theory
- loss aversion
= treat losses more seriously than gains of the same amount (care more about losing than winning a dollar)
- loss neutrality = participants do not favour one option over other unless stakes are high
bounded rationality
- elimination-by-aspects theory
= decision makers eliminate options by considering one relevant attribute/aspect after another
√ useful at early stages
elimination-by-aspects theory
- two-stage theory
= modified version of elimination-by-aspect
- initial stage: elimination-by-aspects –> only options meeting certain criteria are retained
- reduce options to manageable number - second stage: detailed comparisons of patterns of attributes of retained options
- -> few options switching to more complex strategy (multi-attribute utility theory)
different search strategies
- focused search: focus on small number of carefully selected potential options
a. correlated positively with satisfaction with option - exploratory search: taking into account several options making use of several information sources
- haphazard search: non-strategic approach; resembles trial and error
a. correlated negatively
unconscious thought theory
= conscious thinking is constrained by limited capacity of consciousness; unconscious thinking is better at integrating large amounts of information
- conscious thought can follow strict rules (well suited to rule-based problems)
- unconscious though better when decision complex
- optimal: combination of conscious + unconscious thought
x inconsistent findings
x rely heavily on intuitive processes
