Judgement & Decision Making Flashcards
define decision making
choosing a course of action
define judgement
making an evaluation
define reasoning
drawing out further conclusions from a set of facts
what is the normative question?
how should we think?
what are common normative frameworks?
expected utility theory weighted additive model eg lens model; logic eg propositional logic; probability eg. Bayes theorem
what does a ‘normative’ framework’ calculate?
the optimal solution - the definition of being rational
what is the equation for expected utility?
likelihood x benefit
describe the expected utility theory
can you do the calculation with factors that are hard to quantity? these have the more abstract notion of ‘utility’ ie. how much good would each one provide - subjective expected utility theory
what is the descriptive question?
how do we actually think?
what is a cognitive bias?
a systematic deviation from the normative solution: not a random error but a specific pattern of non-optimal behaviour that people repeatedly show
name some examples of biases
confirmation; hyperbolic discounting; illusory truth effect; sunk cost effect; disposition effect; IKEA effect
describe confirmation bias
Nickerson, 1998. tendency to focus on info that supports your opinion
describe hyperbolic discounting
Kirby & Hernstein, 1995. prefer immediate to long term payoffs
describe the illusory truth effect
Begg, Annas, Farinacci, 1992. repeated info more likely to be judged as true.
describe the sunk cost effect
Arkes & Blumer, 1985. continue a behaviour because of previous investment when no longer sensible to do so
describe the disposition effect
Shefrin & Statman, 1985. sell shares that have gone up, keep shares that have gone down
describe the IKEA effect
Norton, Motion & Ariely, 2012. prefer self-made products, meals etc.
describe prospect theory
cognitive biases show we do not calculate expected utility accurately. p.t describes somatic biases in our valuation of likelihood and benefits
describe how we evaluate whether an outcome is good or bad
with reference to our current state, however relatively good or bad that might be. we consider losses as more serious than equivalent gains, and there are diminishing returns on the benefits we gain from positive outcomes or loss from negative outcomes
describe overweight certainty
the certainty effect. guaranteeing something will occur is in itself useful
describe overweight small probabilities
eg. i might win the lottery
describe underweight probable but not certain events
eg what if we get a hard exam q
describe the framing effect
presenting information as a gain or loss causes people to be risk seeking or risk averse
how do traditional economic models assume people act?
rationally - they don’t
describe the status quo bias
Samuelson & Zeckhauser, 1988. tendency to prefer things to stay the same. not rational. leads to policy of defaults eg credit cards
describe the lens model
- identity all the cues. 2. assess weights for each cue. 3. add up the weighted cues
what are heuristics?
mental shortcuts that reduce cognitive effort when thinking. work well in most cases but prone to biases
describe the availability heuristic
how easy it is to think of example is used as a cue to judge likelihood. often reasonable: easier to think of example of events that are frequent. but can be based: infrequent but viv events not a good measure of likelihood
describe the representativeness heuristic
similarity to a category is used to judge likelihood. eg. Linda problem
describe the anchoring & adjustment heuristic
initial anchor used to make the judgement, and then adjustments are made. eg. did Gandhi die before or after the age of 9 (mean estimate 50) /140 (mean estimate 67) ?. both anchors clearly wrong, but people do not adjust sufficiently
describe fast & frugal heuristics
simple processes, using little information and do little integrating. ecologically rational: exploits typical environmental structures
describe the recognition heuristic
eg. does san antonio or san diego have a bigger pop? most say san diego as it more well known. makes sense. simple heuristics that are adapted to the environment can be very effective
describe the fast & frugal tree
Green & Mehr (1997). Limited number of cues. simple processing (yes/no). fit data as well as more complex models. robust to changes in the environment
what is affect?
a general term for mood and emotion. valence & arousal
what is mood?
diffuse feelings not linked to a specific thing
what is emotion?
more specific, differentiated state (eg.fear/anger etc all different). linked to a specific thing eg. a person to event
describe priming in regards to semantic memory models
mood primes recall fo congruent events. bad mood enhances perceived risk of fire, flood etc, and vice versa, even when the mood is not related to the vent. availability heuristic.
describe the affect-as-information hypothesis
affective cues are used directly to evaluate the object of judgement.. but can misattribute the source of feelings. judgements of life satisfaction for on rainy days. effect disappeared when asked about the weather (Schwarz & Clore, 1983)
describe the somatic marker hypothesis
through experience, events are associated with positive/negative feelings that are associated with bodily states. when situ encountered again, feeling is evoked serving as alarm or incentive
describe the Iowa gambling task
bad decks (large rewards but long term loss) and good decks (small rewards but long term gains). repeatedly select cards from decks and receive gain/loss. normal ps learn to choose from good deck. ps with damage to VMPFC (relates emotions) don’t learn - insensitive to future negative consequences) - Bechara & colleagues, 1994
describe the affect heuristic
through experiences, events associated with good/bad outcomes, and representations tagged good/bad. to judge risk/benefits, refer to ‘pool’ of tags and form overall affective judgement. risk/benefit often seen as negative correlated, even when they may not be. this correlation increases under time pressure - more use of heuristic than analytic thinking
which is more specific, emotions or mood?
emotions
emotions we a similar valence can have different…
effects
give an example of emotions with similar valence having different effects
fear = uncertainty and situational control. anger = certainty and individual control
describe naturalistic decision making
suggests all theories so far do not generalise to ‘real world’ decision making.
what are the problems that naturalistic decision making that attempts to overcome?
decision maker expertise, ill-structured problems, incomplete info, shifting goals, decision re multiple events with feedback loops, time constraints, high stakes, organisational norms
describe expert decisions
extensive domain knowledge. stored in templates which are used to identify typical scenarios. templates also allow more complex analysis of consequences
describe recognition primed decision making
model of expert decision making. began with a study of fire commanders. expected people to compare two options. claimed not to make decisions at all
describe 1st RPD strategy: situation assessment
assess situate to establish current goals, cues and expectations. match to prototypical situations (composites of experiences and training). primes known course of action. rapid, intuitive
describe 2nd RPD strategy: diagnosis
no clear match to prototype? match to more than 1? violate expectations? seek info to fill gaps. guided by expectations. develop better SA so that decision is primed
describe 3rd RPD strategy: mental simulation
not encountered situation before? rely on knowledge of elements in the environment. run mental simulation of a course of action
describe explanation-based DM
Pennington & Hastie (1988). developed from jury decisions. people construct causal model of what occurred using evidence/general knowledge. this intermediary representation is basis for decision rather than raw evidence
what are the 3 factors that influence explanation construction according to explanation-based DM?
the evidence; general knowledge about similar events; general knowledge about what constitutes a complete explanation (expect goals, actions and consequences)
what is used to augment explanation according to explanation-based DM?
general knowledge - infer missing details and exclude irrelevant info
in regards to constructing the explanation, people have different general knowledge about… according to explanation-based DM
motives and explanations of behaviour
in regards to constructing the explanation, one is favoured on the ground of what according to explanation-based DM?
coherence: completeness, consistency, plausibility
how is a decision made according to explanation-based DM?
based on match of story to alternative in set of choices.
what leads to variability in decisions according to explanation-based DM?
variability in stories
give some evidence for explanation-based DM
after jury task, ps more likely to recognise real&false evidence if they were consistent with the verdict they gave - implies they construct/embellish story model. Pennington & Hastie, 1988
